1 ## <a name="3 AFS administration"></a> 3 AFS administration
3 The Administration Section of the [[AFSFrequentlyAskedQuestions]].
11 <li><a href="#3 AFS administration"> 3 AFS administration</a><ul>
12 <li><a href="#3.01 Is there a version of xdm"> 3.01 Is there a version of xdm available with AFS authentication?</a></li>
13 <li><a href="#3.02 Is there a version of xloc"> 3.02 Is there a version of xlock available with AFS authentication?</a></li>
14 <li><a href="#3.03 What is /afs/@cell?"> 3.03 What is /afs/@cell?</a></li>
15 <li><a href="#3.04 Given that AFS data is loc"> 3.04 Given that AFS data is location independent, how does an AFS client determine which server houses the data its user is attempting to access?</a></li>
16 <li><a href="#3.05 How does AFS maintain cons"> 3.05 How does AFS maintain consistency on read-write files?</a></li>
17 <li><a href="#3.06 Which protocols does AFS u"> 3.06 Which protocols does AFS use?</a></li>
18 <li><a href="#3.07 Which TCP/IP ports and pro"> 3.07 Which TCP/IP ports and protocols do I need to enable in order to operate AFS through my Internet firewall?</a></li>
19 <li><a href="#3.08 Are setuid programs execut"> 3.08 Are setuid programs executable across AFS cell boundaries?</a></li>
20 <li><a href="#3.09 How can I run daemons with"> 3.09 How can I run daemons with tokens that do not expire?</a></li>
21 <li><a href="#3.10 Can I check my user's pass"> 3.10 Can I check my user's passwords for security purposes?</a></li>
22 <li><a href="#3.11 Is there a way to automati"> 3.11 Is there a way to automatically balance disk usage across fileservers?</a></li>
23 <li><a href="#3.12 Can I shutdown an AFS file"> 3.12 Can I shutdown an AFS fileserver without affecting users?</a></li>
24 <li><a href="#3.13 How can I set up mail deli"> 3.13 How can I set up mail delivery to users with $HOMEs in AFS?</a></li>
25 <li><a href="#3.14 Should I replicate a _Read"> 3.14 Should I replicate a ReadOnly volume on the same partition and server as the ReadWrite volume?</a></li>
26 <li><a href="#3.15 Should I start AFS before"> 3.15 Should I start AFS before NFS in /etc/inittab?</a></li>
27 <li><a href="#3.16 Will AFS run on a multi-ho"> 3.16 Will AFS run on a multi-homed fileserver?</a></li>
28 <li><a href="#3.17 Can I replicate my user's"> 3.17 Can I replicate my user's home directory AFS volumes?</a></li>
29 <li><a href="#3.18 What is the Andrew Benchma"> 3.18 What is the Andrew Benchmark?</a></li>
30 <li><a href="#3.19 Where can I find the Andre"> 3.19 Where can I find the Andrew Benchmark?</a></li>
31 <li><a href="#3.20 Is there a version of HP V"> 3.20 Is there a version of HP VUE login with AFS authentication?</a></li>
32 <li><a href="#3.21 How can I list which clien"> 3.21 How can I list which clients have cached files from a server?</a></li>
33 <li><a href="#3.22 Do Backup volumes require"> 3.22 Do Backup volumes require as much space as ReadWrite volumes?</a></li>
34 <li><a href="#3.23 Should I run timed on my A"> 3.23 Should I run timed on my AFS client?</a></li>
35 <li><a href="#3.24 Why should I keep /usr/vic"> 3.24 Why should I keep /usr/vice/etc/CellServDB current?</a></li>
36 <li><a href="#3.25 How can I keep /usr/vice/e"> 3.25 How can I keep /usr/vice/etc/CellServDB current?</a></li>
37 <li><a href="#3.26 How can I compute a list o"> 3.26 How can I compute a list of AFS fileservers?</a></li>
38 <li><a href="#3.27 How can I set up anonymous"> 3.27 How can I set up anonymous FTP login to access /afs?</a></li>
39 <li><a href="#3.28 Is the data sent over the n"> 3.28 Is the data sent over the network encrypted in AFS ?</a></li>
40 <li><a href="#3.29 What underlying filesystems"> 3.29 What underlying filesystems can I use for AFS ?</a></li>
41 <li><a href="#3.30 Compiling _OpenAFS"> 3.30 Compiling OpenAFS</a></li>
42 <li><a href="#3.31 Upgrading _OpenAFS"> 3.31 Upgrading OpenAFS</a></li>
43 <li><a href="#3.32 Debugging _OpenAFS"> 3.32 Debugging OpenAFS</a></li>
44 <li><a href="#3.33 Tuning client cache for hug"> 3.33 Tuning client cache for huge data</a></li>
45 <li><a href="#3.34 Settting up PAM with AFS"> 3.34 Settting up PAM with AFS</a></li>
46 <li><a href="#3.35 Setting up AFS key in KDC a"> 3.35 Setting up AFS key in KDC and KeyFile</a></li>
47 <li><a href="#3.36 Obtaining and getting asetk"> 3.36 Obtaining and getting asetkey compiled</a></li>
48 <li><a href="#3.37 afs_krb_get_lrealm() using"> 3.37 afs_krb_get_lrealm() using /usr/afs/etc/krb.conf</a></li>
49 <li><a href="#3.38 Moving from kaserver to Hei"> 3.38 Moving from kaserver to Heimdal KDC</a></li>
50 <li><a href="#3.39 Moving from KTH-KRB4 to Hei"> 3.39 Moving from KTH-KRB4 to Heimdal KDC</a></li>
51 <li><a href="#3.40 What are those bos(1) -type"> 3.40 What are those bos(1) -type values simple and cron?</a></li>
52 <li><a href="#3.41 KDC listens on port 88 inst"> 3.41 KDC listens on port 88 instead of 750</a></li>
53 <li><a href="#3.42 afsd gives me "Error -1 in"> 3.42 afsd gives me "Error -1 in basic initialization." on startup</a></li>
54 <li><a href="#3.43 Although I get krb tickets,"> 3.43 Although I get krb tickets, afslog doesn't give me tokens, I see UDP packets to port 4444</a></li>
55 <li><a href="#3.44 I get error message trhough"> 3.44 I get error message trhough syslogd: "afs: Tokens for user of AFS id 0 for cell foo.bar.baz are discarded (rxkad error=19270407)"</a></li>
56 <li><a href="#3.45 I get tickets and tokens, b"> 3.45 I get tickets and tokens, but still get Permission denied.</a></li>
57 <li><a href="#3.46 Recovering broken afs cache"> 3.46 Recovering broken afs cache on clients</a></li>
58 <li><a href="#3.47 What does it mean for a vol"> 3.47 What does it mean for a volume to not be in the VLDB?</a></li>
59 <li><a href="#3.48 What is a Volume Group?"> 3.48 What is a Volume Group?</a></li>
60 <li><a href="#3.49 What is a Clone?"> 3.49 What is a Clone?</a></li>
61 <li><a href="#3.50 What is a Shadow?"> 3.50 What is a Shadow?</a></li>
62 <li><a href="#3.51 Can I authenticate to my af"> 3.51 Can I authenticate to my afs cell using multiple kerberos Realms?</a></li>
63 <li><a href="#3.52 Is it a good idea to store"> 3.52 Is it a good idea to store mail in AFS?</a></li>
64 <li><a href="#3.53 How can I ensure that the"> 3.53 How can I ensure that the userids on client machines match the users' pts ids?</a></li>
65 <li><a href="#3.54 What is Fast Restart?"> 3.54 What is Fast Restart?</a></li>
66 <li><a href="#3.55 Reboot ?"> 3.55 Why does AFS reboot itself spontaneously at 4:00am every Sunday?</a></li>
67 <li><a href="#3.56 weak crypto">3.56 Why do I get an error -1765328370 when authenticating?</a></li>
78 ### <a name="3.01 Is there a version of xdm"></a><a name="3.01 Is there a version of xdm "></a> 3.01 Is there a version of xdm available with AFS authentication?
80 These days this kind of functionality is provided by something like PAM.
81 Acquiring AFS tokens via PAM can be done by a few different PAM modules,
82 including Russ Allbery's
83 [pam-afs-session](http://www.eyrie.org/~eagle/software/pam-afs-session/).
85 ### <a name="3.02 Is there a version of xloc"></a> 3.02 Is there a version of xlock available with AFS authentication?
87 See previous question.
89 ### <a name="3.03 What is /afs/@cell?"></a> 3.03 What is /afs/@cell?
91 It is a symbolic link pointing at /afs/$your\_cell\_name.
93 NB, @cell is not something that is provided by AFS. You may decide it is useful in your cell and wish to create it yourself.
95 /afs/@cell is useful because:
97 - If you look after more than one AFS cell, you could create the link in each cell then set your PATH as:
98 - PATH=$PATH:/afs/@cell/@sys/local/bin
100 - For most cells, it shortens the path names to be typed in thus reducing typos and saving time.
102 A disadvantage of using this convention is that when you cd into /afs/@cell then type "pwd" you see "/afs/@cell" instead of the full name of your cell. This may appear confusing if a user wants to tell a user in another cell the pathname to a file.
104 You could create your own /afs/@cell with the following:
108 [ -L /afs/@cell ] && echo We already have @cell! && exit
109 cell=$(cat /usr/vice/etc/ThisCell)
110 cd /afs/.${cell} && fs mkm temp root.afs
112 ln -s /afs/${cell} @cell
113 ln -s /afs/.${cell} .@cell # .@cell for RW path
114 cd /afs/.${cell} && fs rmm temp
115 vos release root.afs; fs checkv
117 <http://www-archive.stanford.edu/lists/info-afs/hyper95/0298.html>
119 ### <a name="3.04 Given that AFS data is loc"></a> 3.04 Given that AFS data is location independent, how does an AFS client determine which server houses the data its user is attempting to access?
121 The Volume Location Database (VLDB) is stored on AFS Database Servers and is ideally replicated across 3 or more Database Server machines. Replication of the Database ensures high availability and load balances the requests for the data. The VLDB maintains information regarding the current physical location of all volume data (files and directories) in the cell, including the IP address of the [[FileServer]], and the name of the disk partition the data is stored on.
123 A list of a cell's Database Servers is stored on the local disk of each AFS Client machine as: /usr/vice/etc/CellServDB
125 The Database Servers also house the Kerberos Authentication Database (encrypted user and server passwords), the Protection Database (user UID and protection group information) and the Backup Database (used by System Administrators to backup AFS file data to tape).
127 ### <a name="3.05 How does AFS maintain cons"></a> 3.05 How does AFS maintain consistency on read-write files?
129 AFS uses a mechanism called "callback".
131 Callback is a promise from the fileserver that the cache version of a file/directory is up-to-date. It is established by the fileserver with the caching of a file.
133 When a file is modified the fileserver breaks the callback. When the user accesses the file again the Cache Manager fetches a new copy if the callback has been broken.
135 The following paragraphs describe AFS callback mechanism in more detail:
137 If I open() fileA and start reading, and you then open() fileA, write() a change **\*\*and close() or fsync()\*\*** the file to get your changes back to the server - at the time the server accepts and writes your changes to the appropriate location on the server disk, the server also breaks callbacks to all clients to which it issued a copy of fileA.
139 So my client receives a message to break the callback on fileA, which it dutifully does. But my application (editor, spreadsheet, whatever I'm using to read fileA) is still running, and doesn't really care that the callback has been broken.
141 When something causes the application to read() more of the file the read() system call executes AFS cache manager code via the VFS switch, which does check the callback and therefore gets new copies of the data.
143 Of course, the application may not re-read data that it has already read, but that would also be the case if you were both using the same host. So, for both AFS and local files, I may not see your changes.
145 Now if I exit the application and start it again, or if the application does another open() on the file, then I will see the changes you've made.
147 This information tends to cause tremendous heartache and discontent - but unnecessarily so. People imagine rampant synchronization problems. In practice this rarely happens and in those rare instances, the data in question is typically not critical enough to cause real problems or crashing and burning of applications. Since 1985, we've found that the synchronization algorithm has been more than adequate in practice - but people still like to worry!
149 The source of worry is that, if I make changes to a file from my workstation, your workstation is not guaranteed to be notified until I close or fsync the file, at which point AFS guarantees that your workstation will be notified. This is a significant departure from NFS, in which no guarantees are provided.
151 Partially because of the worry factor and largely because of Posix, this will change in DFS. DFS synchronization semantics are identical to local file system synchronization.
153 [ DFS is the Distributed File System which is part of the Distributed ] [ Computing Environment (DCE). ]
155 ### <a name="3.06 Which protocols does AFS u"></a> 3.06 Which protocols does AFS use?
157 AFS may be thought of as a collection of protocols and software processes, nested one on top of the other. The constant interaction between and within these levels makes AFS a very sophisticated software system.
159 At the lowest level is the UDP protocol. UDP is the connection to the actual network wire. The next protocol level is the remote procedure call (RPC). In general, RPCs allow the developer to build applications using the client/server model, hiding the underlying networking mechanisms. AFS uses Rx, an RPC protocol developed specifically for AFS during its development phase at Carnegie Mellon University.
161 Above the RPC is a series of server processes and interfaces that all use Rx for communication between machines. Fileserver, volserver, upserver, upclient, and bosserver are server processes that export RPC interfaces to allow their user interface commands to request actions and get information. For example, a bos status command will examine the bos server process on the indicated file server machine.
163 Database servers use ubik, a replicated database mechanism which is implemented using RPC. Ubik guarantees that the copies of AFS databases of multiple server machines remain consistent. It provides an application programming interface (API) for database reads and writes, and uses RPCs to keep the database synchronized. The database server processes, vlserver, kaserver, and ptserver, reside above ubik. These processes export an RPC interface which allows user commands to control their operation. For instance, the pts command is used to communicate with the ptserver, while the command klog uses the kaserver's RPC interface.
165 Some application programs are quite complex, and draw on RPC interfaces for communication with an assortment of processes. Scout utilizes the RPC interface to file server processes to display and monitor the status of file servers. The uss command interfaces with kaserver, ptserver, volserver and vlserver to create new user accounts.
167 The Cache Manager also exports an RPC interface. This interface is used principally by file server machines to break callbacks. It can also be used to obtain Cache Manager status information. The program cmdebug shows the status of a Cache Manager using this interface.
169 For additional information, Section 1.5 of the AFS System Administrator's Guide and the April 1990 Cache Update contain more information on ubik. Udebug information and short descriptions of all debugging tools were included in the January 1991 Cache Update. Future issues will discuss other debugging tools in more detail.
171 [ source: <ftp://ftp.transarc.com/pub/afsug/newsletter/apr91> ] [ Copyright 1991 Transarc Corporation ]
173 ### <a name="3.07 Which TCP/IP ports and pro"></a> 3.07 Which TCP/IP ports and protocols do I need to enable in order to operate AFS through my Internet firewall?
175 Assuming you have already taken care of nameserving, you may wish to use an Internet timeserver for Network Time Protocol [[[NTP|FurtherReading#NTP]]] and the question about timed:
179 A list of NTP servers is available via anonymous FTP from:
181 - <http://www.eecis.udel.edu/~mills/ntp/servers.html>
183 For further details on NTP see: <http://www.eecis.udel.edu/~ntp/>
185 For a "minimal" AFS service which does not allow inbound or outbound klog:
187 cachemanager 4711/udp (only if you use the arla-client instead of OpenAFS)
189 cachemanager 7001/udp
197 (Ports in the 7020-7029 range are used by the AFS backup system, and won't be needed by external clients performing simple file accesses.)
199 Additionally, for "klog" to work through the firewall you need to allow inbound and outbound UDP on ports >1024 (probably 1024<port<2048 would suffice depending on the number of simultaneous klogs).
201 See also: <http://www-archive.stanford.edu/lists/info-afs/hyper95/0874.html>
203 ### <a name="3.08 Are setuid programs execut"></a> 3.08 Are setuid programs executable across AFS cell boundaries?
205 By default, the setuid bit is ignored but the program may be run (without setuid privilege).
207 It is possible to configure an AFS client to honour the setuid bit. This is achieved by root running:
209 root@toontown # fs setcell -cell $cellname -suid
211 (where $cellname is the name of the foreign cell. Use with care!).
213 NB: making a program setuid (or setgid) in AFS does **not** mean that the program will get AFS permissions of a user or group. To become AFS authenticated, you have to klog. If you are not authenticated, AFS treats you as "system:anyuser".
215 ### <a name="3.09 How can I run daemons with"></a> 3.09 How can I run daemons with tokens that do not expire?
217 It is not a good idea to run with tokens that do not expire because this would weaken one of the security features of Kerberos.
219 A better approach is to re-authenticate just before the token expires.
221 The most common way to achieve this these days is to generate a keytab
222 containing the credentials you want the daemon to have, and use a program like
223 [k5start](http://www.eyrie.org/~eagle/software/kstart/) to run the daemon with
224 those credentials. The kstart tools can be compared to the old "reauth"
227 ### <a name="3.10 Can I check my user's pass"></a> 3.10 Can I check my user's passwords for security purposes?
229 Yes. Alec Muffett's Crack tool (at version 4.1f) has been converted to work on the Transarc kaserver database. This modified Crack (AFS Crack) is available via anonymous ftp from:
231 - <ftp://export.acs.cmu.edu/pub/crack.tar.Z>
232 - <ftp://ftp.andrew.cmu.edu/pub/AFS-Tools/crack.tar.Z>
234 and is known to work on: pmax\_\* sun4\*\_\* hp700\_\* rs\_aix\* next\_\*
236 It uses the file /usr/afs/db/kaserver.DB0, which is the database on the kaserver machine that contains the encrypted passwords. As a bonus, AFS Crack is usually two to three orders of magnitude faster than the standard Crack since there is no concept of salting in a Kerberos database.
238 On a normal UNIX /etc/passwd file, each password can have been encrypted around 4096 (2^12) different saltings of the crypt(3) algorithm, so for a large number of users it is easy to see that a potentially large (up to 4095) number of seperate encryptions of each word checked has been avoided.
240 Author: Dan Lovinger Contact: Derrick J. Brashear <shadow+@andrew.cmu.edu>
242 <table border="1" cellpadding="0" cellspacing="0">
245 <td> AFS Crack does not work for MIT Kerberos Databases. The author is willing to give general guidance to someone interested in doing the (probably minimal) amount of work to port it to do MIT Kerberos. The author does not have access to a MIT Kerberos server to do this. </td>
249 ### <a name="3.11 Is there a way to automati"></a> 3.11 Is there a way to automatically balance disk usage across fileservers?
251 Yes. There is a tool, balance, which does exactly this. It can be retrieved via anonymous ftp from:
253 - <ftp://ftp.andrew.cmu.edu/pub/AFS-Tools/balance-1.1b.tar.gz>
255 Actually, it is possible to write arbitrary balancing algorithms for this tool. The default set of "agents" provided for the current version of balance balance by usage, # of volumes, and activity per week, the latter currently requiring a source patch to the AFS volserver. Balance is highly configurable.
257 Author: Dan Lovinger Contact: Derrick Brashear <shadow+@andrew.cmu.edu>
259 ### <a name="3.12 Can I shutdown an AFS file"></a> 3.12 Can I shutdown an AFS fileserver without affecting users?
261 Yes, this is an example of the flexibility you have in managing AFS.
263 Before attempting to shutdown an AFS fileserver you have to make some arrangements that any services that were being provided are moved to another AFS fileserver:
265 1. Move all AFS volumes to another fileserver. (Check you have the space!) This can be done "live" while users are actively using files in those volumes with no detrimental effects.
267 1. Make sure that critical services have been replicated on one (or more) other fileserver(s). Such services include:
268 - kaserver - Kerberos Authentication server
269 - vlserver - Volume Location server
270 - ptserver - Protection server
271 - buserver - Backup server
273 It is simple to test this before the real shutdown by issuing:
275 bos shutdown $server $service
277 where: $server is the name of the server to be shutdown
278 and $service is one (or all) of: kaserver vlserver ptserver buserver
280 Other points to bear in mind:
282 - "vos remove" any RO volumes on the server to be shutdown. Create corresponding RO volumes on the 2nd fileserver after moving the RW. There are two reasons for this:
283 1. An RO on the same partition ("cheap replica") requires less space than a full-copy RO.
284 2. Because AFS always accesses RO volumes in preference to RW, traffic will be directed to the RO and therefore quiesce the load on the fileserver to be shutdown.
286 - If the system to be shutdown has the lowest IP address there may be a brief delay in authenticating because of timeout experienced before contacting a second kaserver.
288 ### <a name="3.13 How can I set up mail deli"></a> 3.13 How can I set up mail delivery to users with $HOMEs in AFS?
290 There are many ways to do this. Here, only two methods are considered:
292 Method 1: deliver into local filestore
294 This is the simplest to implement. Set up your mail delivery to append mail to /var/spool/mail/$USER on one mailserver host. The mailserver is an AFS client so users draw their mail out of local filestore into their AFS $HOME (eg: inc).
296 Note that if you expect your (AFS unauthenticated) mail delivery program to be able to process .forward files in AFS $HOMEs then you need to add "system:anyuser rl" to each $HOMEs ACL.
300 - Simple to implement and maintain.
301 - No need to authenticate into AFS.
305 - It doesn't scale very well.
306 - Users have to login to the mailserver to access their new mail.
307 - Probably less secure than having your mailbox in AFS.
308 - System administrator has to manage space in /var/spool/mail.
310 Method 2: deliver into AFS
312 This takes a little more setting up than the first method.
314 First, you must have your mail delivery daemon AFS authenticated (probably as "postman"). The reauth example in afs-contrib shows how a daemon can renew its token. You will also need to setup the daemon startup soon after boot time to klog (see the -pipe option).
316 Second, you need to set up the ACLs so that "postman" has lookup rights down to the user's $HOME and "lik" on $HOME/Mail.
320 - Scales better than first method.
321 - Delivers to user's $HOME in AFS giving location independence.
322 - Probably more secure than first method.
323 - User responsible for space used by mail.
327 - More complicated to set up.
328 - Need to correctly set ACLs down to $HOME/Mail for every user.
329 - Probably need to store postman's password in a file so that the mail delivery daemon can klog after boot time. This may be OK if the daemon runs on a relatively secure host.
331 An example of how to do this for IBM RISC System/6000 is auth-sendmail. A beta test version of auth-sendmail can be found in:
333 <file:///afs/transarc.com/public/afs-contrib/doc/faq/auth-sendmail.tar.Z> <ftp://ftp.transarc.com/pub/afs-contrib/doc/faq/auth-sendmail.tar.Z>
335 ### <a name="3.14 Should I replicate a _Read"></a> 3.14 Should I replicate a [[ReadOnly]] volume on the same partition and server as the [[ReadWrite]] volume?
337 Yes, Absolutely! It improves the robustness of your served volumes.
339 If [[ReadOnly]] volumes exist (note use of term **exist** rather than **are available**), Cache Managers will **never** utilize the [[ReadWrite]] version of the volume. The only way to access the RW volume is via the "dot" path (or by special mounting).
341 This means if **all** RO copies are on dead servers, are offline, are behind a network partition, etc, then clients will not be able to get the data, even if the RW version of the volume is healthy, on a healthy server and in a healthy network.
343 However, you are **very** strongly encouraged to keep one RO copy of a volume on the **same server and partition** as the RW. There are two reasons for this:
345 1. The RO that is on the same server and partition as the RW is a clone (just a copy of the header - not a full copy of each file). It therefore is very small, but provides access to the same set of files that all other (full copy) [[ReadOnly]] volume do. Transarc trainers refer to this as the "cheap replica".
346 2. To prevent the frustration that occurs when all your ROs are unavailable but a perfectly healthy RW was accessible but not used.
348 If you keep a "cheap replica", then by definition, if the RW is available, one of the RO's is also available, and clients will utilize that site.
350 ### <a name="3.15 Should I start AFS before"></a><a name="3.15 Should I start AFS before "></a> 3.15 Should I start AFS before NFS in /etc/inittab?
352 Yes, it is possible to run both AFS and NFS on the same system but you should start AFS first.
354 In IBM's AIX 3.2, your /etc/inittab would contain:
356 rcafs:2:wait:/etc/rc.afs > /dev/console 2>&1 # Start AFS daemons
357 rcnfs:2:wait:/etc/rc.nfs > /dev/console 2>&1 # Start NFS daemons
359 With AIX, you need to load NFS kernel extensions before the AFS KEs in /etc/rc.afs like this:
362 # example /etc/rc.afs for an AFS fileserver running AIX 3.2
364 echo "Installing NFS kernel extensions (for AFS+NFS)"
365 /etc/gfsinstall -a /usr/lib/drivers/nfs.ext
366 echo "Installing AFS kernel extensions..."
367 D=/usr/afs/bin/dkload
368 ${D}/cfgexport -a ${D}/export.ext
369 ${D}/cfgafs -a ${D}/afs.ext
370 /usr/afs/bin/bosserver &
372 ### <a name="3.16 Will AFS run on a multi-ho"></a> 3.16 Will AFS run on a multi-homed fileserver?
374 (multi-homed = host has more than one network interface.)
376 Yes, it will. However, AFS was designed for hosts with a single IP address. There can be problems if you have one host name being resolved to several IP addresses.
378 Transarc suggest designating unique hostnames for each network interface. For example, a host called "spot" has two tokenring and one ethernet interfaces: spot-tr0, spot-tr1, spot-en0. Then, select which interface will be used for AFS and use that hostname in the [[CellServDB]] file (eg: spot-tr0).
380 You also have to remember to use the AFS interface name with any AFS commands that require a server name (eg: vos listvol spot-tr0).
382 There is a more detailed discussion of this in the August 1993 issue of "Cache Update" (see: <ftp://ftp.transarc.com/pub/afsug/newsletter/aug93>).
384 The simplest way of dealing with this is to make your AFS fileservers single-homed (eg only use one network interface).
386 At release 3.4 of AFS, it is possible to have multi-homed fileservers (but _not_ multi-homed database servers).
388 ### <a name="3.17 Can I replicate my user's"></a><a name="3.17 Can I replicate my user's "></a> 3.17 Can I replicate my user's home directory AFS volumes?
392 Users with $HOMEs in /afs normally have an AFS [[ReadWrite]] volume mounted in their home directory.
394 You can replicate a RW volume but only as a [[ReadOnly]] volume and there can only be one instance of a [[ReadWrite]] volume.
396 In theory, you could have RO copies of a user's RW volume on a second server but in practice this won't work for the following reasons:
398 a) AFS has built-in bias to always access the RO copy of a RW volume.
399 So the user would have a ReadOnly $HOME which is not too useful!
401 b) Even if a) was not true you would have to arrange frequent
402 synchronisation of the RO copy with the RW volume (for example:
403 "vos release user.fred; fs checkv") and this would have to be
404 done for all such user volumes.
406 c) Presumably, the idea of replicating is to recover the $HOME
407 in the event of a server crash. Even if a) and b) were not
408 problems consider what you might have to do to recover a $HOME:
410 1) Create a new RW volume for the user on the second server
411 (perhaps named "user.fred.2").
413 2) Now, where do you mount it?
415 The existing mountpoint cannot be used because it already has
416 the ReadOnly copy of the original volume mounted there.
418 Let's choose: /afs/MyCell/user/fred.2
420 3) Copy data from the RO of the original into the new RW volume
423 4) Change the user's entry in the password file for the new $HOME:
424 /afs/MyCell/user/fred.2
426 You would have to attempt steps 1 to 4 for every user who had
427 their RW volume on the crashed server. By the time you had done
428 all of this, the crashed server would probably have rebooted.
430 The bottom line is: you cannot replicate $HOMEs across servers.
432 ### <a name="3.18 What is the Andrew Benchma"></a> 3.18 What is the Andrew Benchmark?
434 "It is a script that operates on a collection of files constituting an application program. The operations are intended to represent typical actions of an average user. The input to the benchmark is a source tree of about 70 files. The files total about 200 KB in size. The benchmark consists of five distinct phases:
436 I MakeDir - Construct a target subtree that is identical to the
438 II Copy - Copy every file from the source subtree to the target subtree.
439 III ScanDir - Traverse the target subtree and examine the status
441 IV ReadAll - Scan every byte of every file in the target subtree.
442 V Make - Complete and link all files in the target subtree."
444 Source: <file:///afs/transarc.com/public/afs-contrib/doc/benchmark/Andrew.Benchmark.ps> <ftp://ftp.transarc.com/pub/afs-contrib/doc/benchmark/Andrew.Benchmark.ps>
446 ### <a name="3.19 Where can I find the Andre"></a> 3.19 Where can I find the Andrew Benchmark?
448 From Daniel Joseph Barnhart Clark's [message](http://lists.openafs.org/pipermail/openafs-info/2004-April/012942.html):
450 It's linked to from <http://www.citi.umich.edu/projects/linux-scalability/tools.html>; CITI also has some useful mods at <http://www.citi.umich.edu/projects/nfs-perf/results/cmarion/>.
452 ### <a name="3.20 Is there a version of HP V"></a> 3.20 Is there a version of HP VUE login with AFS authentication?
454 Yes, the availability of this is described in: <file:///afs/transarc.com/public/afs-contrib/pointers/HP-VUElogin.txt> <ftp://ftp.transarc.com/pub/afs-contrib/pointers/HP-VUElogin.txt>
456 If you don't have access to the above, please contact Rajeev Pandey of Hewlett Packard whose email address is <rpandey@cv.hp.com>.
458 ### <a name="3.21 How can I list which clien"></a> 3.21 How can I list which clients have cached files from a server?
460 By using the following script:
465 # AUTHOR Rainer Toebbicke <rtb@dxcern.cern.ch>
467 # PURPOSE Display AFS clients which have grabbed files from a server
470 echo "Usage: $0 <afs_server 1> ... <afsserver n>"
474 /usr/afsws/etc/rxdebug -servers $n -allconn
475 done | grep '^Connection' | \
476 while read x y z ipaddr rest; do echo $ipaddr; done | sort -u |
477 while read ipaddr; do
479 n="`nslookup $ipaddr`"
487 ### <a name="3.22 Do Backup volumes require"></a><a name="3.22 Do Backup volumes require "></a> 3.22 Do Backup volumes require as much space as [[ReadWrite]] volumes?
491 The technique used is to create a new volume, where every file in the RW copy is pointed to by the new backup volume. The files don't exist in the BK, only in the RW volume. The backup volume therefore takes up very little space.
493 If the user now starts modifying data, the old copy must not be destroyed.
495 There is a Copy-On-Write bit in the vnode; if the fileserver writes to a vnode with the bit on, it allocates a new vnode for the data and turns off the COW bit. The BK volume hangs onto the old data, and the RW volume slowly splits itself away over time.
497 The BK volume is re-synchronised with the RW next time a "vos backupsys" is run.
499 The space needed for the BK volume is directly related to the size of all files changed in the RW between runs of "vos backupsys".
501 ### <a name="3.23 Should I run timed on my A"></a> 3.23 Should I run timed on my AFS client?
505 <a name="NTP"></a> The AFS Servers make use of NTP [[[NTP|FurtherReading#NTP]]] to synchronise time each other and typically with one or more external NTP servers. By default, clients synchronize their time with one of the servers in the local cell. Thus all the machines participating in the AFS cell have an accurate view of the time.
507 For further details on NTP see: <http://www.eecis.udel.edu/~ntp/>. The latest version is 4.1, dated August 2001, which is **much** more recent that the version packaged with Transarc AFS.
509 A list of NTP servers is available via anonymous FTP from:
511 - <http://www.eecis.udel.edu/~mills/ntp/servers.html>
513 The default time setting behavior of the AFS client can be disabled by specifying the `-nosettime` argument to [afsd](http://www.transarc.ibm.com/Library/documentation/afs/3.5/unix/cmd/cmd53.htm). This is recommended for AFS servers which are also configured as clients (because servers normally run NTP daemons) and for clients that run NTP.
515 ### <a name="3.24 Why should I keep /usr/vic"></a> 3.24 Why should I keep /usr/vice/etc/CellServDB current?
517 On AFS clients, /usr/vice/etc/CellservDB, defines the cells and (their servers) that can be accessed via /afs.
519 Over time, site details change: servers are added/removed or moved onto new network addresses. New sites appear.
521 In order to keep up-to-date with such changes, the [[CellservDB]] file on each AFS client should be kept consistent with some master copy (at your site).
523 As well as updating [[CellservDB]], your AFS administrator should ensure that new cells are mounted in your cell's root.afs volume.
525 If a cell is added to [[CellServDB]] then the **client** must either be restared or you must the AFS command "fs newcell -cell .. -servers ... ...".
527 ### <a name="3.25 How can I keep /usr/vice/e"></a> 3.25 How can I keep /usr/vice/etc/CellServDB current?
529 Do a daily copy from a master source and update the AFS kernel sitelist.
531 One good master source is the grand.central.org [[CellServDB]], available from <http://grand.central.org/dl/cellservdb/CellServDB> or <file:/afs/openafs.org/service/CellServDB> (N.B. update to /afs/grand.central.org path when available). You can send updates for this to <cellservdb@central.org>.
533 The client [[CellServDB]] file must not reside under /afs and is best located in local filespace.
535 Simply updating a client [[CellServDB]] file is not enough. You also need to update the AFS kernel sitelist by either: 1) rebooting the client or 2) running "fs newcell $cell\_name $server\_list" for each site in the [[CellServDB]] file.
537 A script to update the AFS kernel sitelist on a running system is newCellServDB.
539 <file:///afs/ece.cmu.edu/usr/awk/Public/newCellServDB> <ftp://ftp.ece.cmu.edu/pub/afs-tools/newCellServDB>
541 One way to distribute [[CellServDB]] is to have a root cron job on each AFS client copy the file then run newCellServDB.
547 # NAME syncCellServDB
548 # PURPOSE Update local CellServDB file and update AFS kernel sitelist
549 # USAGE run by daily root cron job eg:
550 # 0 3 * * * /usr/local/sbin/syncCellServDB
552 # NOTE "@cell" is a symbolic link to /afs/$this_cell_name
554 src=/afs/@cell/service/etc/CellServDB
555 dst=/usr/vice/etc/CellServDB
556 xec=/usr/local/sbin/newCellServDB
557 log=/var/log/syncCellServDB
559 if [ -s ${src} ]; then
560 if [ ${src} -nt ${dst} ]; then
561 cp $dst ${dst}- && cp $src $dst && $xec 2>&1 >$log
563 echo "master copy no newer: no processing to be done" >$log
566 echo "zero length file: ${src}" >&2
569 ### <a name="3.26 How can I compute a list o"></a> 3.26 How can I compute a list of AFS fileservers?
571 Here is a Bourne shell command to do it (it will work in GNU bash and the Korn shell, too):
573 stimpy@nick $ vos listvldb -cell `cat /usr/vice/etc/ThisCell` \\
574 | awk '(/server/) {print $2}' | sort -u
576 ### <a name="3.27 How can I set up anonymous"></a> 3.27 How can I set up anonymous FTP login to access /afs?
578 The easiest way on a primarily "normal" machine (where you don't want to have everything in AFS) is to actually mount root.cell under ~ftp, and then symlink /afs to ~ftp/afs or whatever. It's as simple as changing the mountpoint in /usr/vice/etc/cacheinfo and restarting afsd.
580 Note that when you do this, anon ftp users can go anywhere system:anyuser can (or worse, if you're using IP-based ACLs and the ftp host is PTS groups). The only "polite" solution I've arrived at is to have the ftp host machine run a minimal [[CellServDB]] and police my ACLs tightly.
582 Alternatively, you can make ~ftp an AFS volume and just mount whatever you need under that - this works well if you can keep everything in AFS, and you don't have the same problems with anonymous "escapes" into /afs.
584 Unless you need to do authenticating ftp, you are _strongly_ recommended using wu-ftpdv2.4 (or better).
586 ### <a name="3.28 Is the data sent over the n"></a> 3.28 Is the data sent over the network encrypted in AFS ?
588 There is still no easy way to do this in Transarc AFS, but [[OpenAFS]] now has a "fs" subcommand to turn on encryption of regular file data sent and received by a client. This is a per client setting that persist until reboot. No server actions are needed to support this change. The syntax is:
594 Note that this only encrypts network traffic between the client and server. The data on the server's disk is not encrypted nor is the data in the client's disk cache. The encryption algorithm used is [fcrypt](http://surfvi.com/~ota/fcrypt-paper.txt), which is a DES variant.
596 Getting encryption enabled by default:
598 - [[RedHat]] Linux: ([src](https://lists.openafs.org/pipermail/openafs-info/2002-July/005085.html)) change the last line of /etc/sysconfig/afs to `AFS_POST_INIT="/usr/bin/fs setcrypt on"`
599 - Windows ([src](https://lists.openafs.org/pipermail/openafs-info/2003-June/009416.html)) set the following registry value named `SecurityLevel` under `HKLM\SYSTEM\CurrentControlSet\Services\TransarcAFSDaemon\Parameters` to 2.
601 I have not tested either of these procedures. -- Ted Anderson - 05 Jun 2003
603 ### <a name="3.29 What underlying filesystems"></a> 3.29 What underlying filesystems can I use for AFS ?
605 See also [[SupportedConfigurations]].
607 You need to distinguish between the filesystem used by the file server to store the actual AFS data (by convention in /vicep?) and the filesystem used by the client cache manager to cache files.
609 Fileserver is started by bosserver. It depends, what ./configure switches were used during compilation from sources. To be always on the safe side, use --enable-namei-fileserver configure flag. That will give you fileserver binary which can act on any /vicep? partition regardless it's filesystem type. With the new namei file server you can basically use any filesystem you want. The namei file server does not do any fancy stuff behind the scenes but only accesses normal files (their names are a bit strange though).
611 The opposite to namei fileserver is inode based nameserver. According to openafs-devel email list, it gave on some Solaris box 10% speedup over the namei based server. The namei based fileserver cannot run on every filesystem, as it stores some internal afs-specific data into the filesystem. Typically, /sbin/fsck distributed withthe operating system zaps these data. Inode based fileserver directly operates on the inodes of the underlying file system which therefore has to fully support the inode abstraction scheme. The the Administrators Guide for more details (they differ from system to system).
613 On the client side where you run /usr/afs/etc/afsd as a kernel process, you always have to use a file system supporting the inode abstraction for the cache (usually /usr/vice/cache) since the cache manager references files by their inode. Fortunately, it does not do such tricky stuff as the inode based fileserver.
615 The following file systems have been reported _not_ to work for the AFS client cache:
619 - advfs (Tru64), it works but gives cache corruption
620 - efs (SGI) - IBM AFS does support efs, but openafs doesn't have a license for that.
622 - Patch committed to cvs around 6/2003 will now enforce this in some cases and generate a warning to the user if the filesystem type is wrong.
624 The following file systems have been reported to work for the AFS client cache:
629 - xfs (at least on IRIX 6.5)
630 - ufs (Solaris, [[Tru64Unix]])
632 ### <a name="3.30 Compiling _OpenAFS"></a> 3.30 Compiling [[OpenAFS]]
634 The kernel parts on Solaris have to be compiled with Sun cc, same for other platforms, i.e. you need same compiler used to compile kernel to compile your afs modules. [[Tru64Unix]] doesn't support modules, so you have to edit kernel config files and link statically into kernel. Kernel module insertion works fine on Linux, Solaris, Irix ...
636 ./configure --enable-transarc-paths=/usr/etc --with-afs-sysname=i386_linux24
640 ... and continue the install process described in IBM AFS documentation. If you do "make install", you will end up with some stuff installed into /usr/local but something not, regardless the --enable-transarc-paths option ... "make install" it's messy.
642 ### <a name="3.31 Upgrading _OpenAFS"></a> 3.31 Upgrading [[OpenAFS]]
644 Upgrade of AFS on Linux
646 /etc/rc.d/init.d/afs stop
647 cd root.client/usr/vice/etc
648 tar cvf - . | (cd /usr/vice/etc; tar xfp -)
649 cp -p afs.rc /etc/rc.d/init.d/afs
650 cp ../../../../lib/pam_afs.krb.so.1 /lib/security
651 cd ../../../../root.server/usr/afs
652 tar cvf - . | (cd /usr/afs; tar xfp -)
653 # echo "auth sufficient /lib/security/pam_afs.so try_first_pass \
654 ignore_root" >> /etc/pam.d/login
656 vim /etc/sysconfig/afs
657 ln -s pam_afs.krb.so.1 pam_afs.so
659 ln -s ../init.d/afs S99afs
661 ln -s ../init.d/afs K01afs
662 cp /usr/vice/etc/afs.conf /etc/sysconfig/afs
663 /etc/rc.d/init.d/afs start
665 Upgrade of AFS on Solaris 2.6
670 cd root.server/usr/afs
671 tar cvf - ./bin | (cd /usr/afs; tar xfp -)
673 cp root.client/usr/vice/etc/modload/libafs.nonfs.o /kernel/fs/afs
674 cp root.server/etc/vfsck /usr/lib/fs/afs/fsck
675 cd root.client/usr/vice
676 tar cvf - ./etc | (cd /usr/vice; tar xf -)
678 cp lib/pam_afs.krb.so.1 /usr/lib/security
679 cp lib/pam_afs.so.1 /usr/lib/security
684 Upgrade of AFS on Irix 6.5
686 /etc/chkconfig -f afsserver off
687 /etc/chkconfig -f afsclient off
688 /etc/chkconfig -f afsml off
689 /etc/chkconfig -f afsxnfs off
691 cd root.server/usr/afs
692 tar cvf - ./bin | (cd /usr/afs; tar xfp -)
694 cp root.client/usr/vice/etc/sgiload/libafs.IP22.nonfs.o /usr/vice/etc/sgiload
695 echo "AFS will be compiled statically into kernel"
696 echo "otherwise skip following lines and use chkconfig afsml on"
697 cp root.client/bin/afs.sm /var/sysgen/system
698 cp root.client/bin/afs /var/sysgen/master.d
699 echo "The next file comes from openafs-*/src/libafs/STATIC.*"
700 cp root.client/bin/libafs.IP22.nonfs.a /var/sysgen/boot/afs.a
703 echo "end of static kernel modifications"
704 cd root.client/usr/vice/etc
705 echo "Delete any of the modload/ files which don't fit your platform if you need space"
706 echo "These files originate from openafs-*/src/libafs/MODLOAD.*"
707 tar cvf - . | (cd /usr/vice/etc; tar xf -)
708 /etc/chkconfig -f afsserver on
709 /etc/chkconfig -f afsclient on
710 # /etc/chkconfig -f afsml on - afs is compiled statically into kernel, so leave afsml off
711 /etc/chkconfig -f afsml off
712 /etc/chkconfig -f afsxnfs off
715 Upgrade of AFS on [[Tru64Unix]]
717 cd root.server/usr/afs/
718 tar cvf - ./bin | (cd /usr/afs; tar xfp -)
719 cd ../../../root.client/bin
720 cp ./libafs.nonfs.o /usr/sys/BINARY/afs.mod
721 ls -la /usr/sys/BINARY/afs.mod
723 cd ../../root.client/usr/vice
724 cp etc/afsd /usr/vice/etc/afsd
725 cp etc/C/afszcm.cat /usr/vice/etc/C/afszcm.cat
727 ### <a name="3.32 Debugging _OpenAFS"></a> 3.32 Debugging [[OpenAFS]]
729 In case of troubles when you need only fileserver process to run (to be able to debug), run the lwp fileserver instead of the pthreads fileserver (src/viced/fileserver instead of src/tviced/fileserver if you have a buildtree handy):
731 cp src/viced/fileserver /usr/afs/bin (or wherever)
733 bos restart calomys fs -local
735 ... then attach with gdb
737 To debug if client running afsd kernel process talks to the servers from [[CellServDB]], do:
739 tcpdump -vv -s 1500 port 7001
743 fileserver 7000/udp # fileport (FILESERVICE)
744 afscb 7001/udp # kernel extensions
745 afsprot 7002/udp # ptserver (PROTSERVICE)
746 afsvldb 7003/udp # vlserver (VLDBSERVICE)
747 afskauth 7004/udp # kaserver (KAUTHSERVICE)
748 afsvol 7005/udp # volserver (VOLUMESERVICE)
749 afserror 7006/udp # ERRORSERVICE
750 afsnanny 7007/udp # bosserver (NANNYSERVICE)
751 afsupdate 7008/udp # upserver (UPDATESERVICE)
752 afsrmtsys 7009/udp # RMTSYSSERVICE
754 When tcpdump doesn't help, try:
756 fstrace setset cm -active; make your error happen; fstrace dump cm
758 ### <a name="3.33 Tuning client cache for hug"></a> 3.33 Tuning client cache for huge data
760 Use on afsd command line -chunk 17 or greater. Be carefull, with certain cache sizes afsd crashes on startup (Linux, [[Tru64Unix]] at least). It is possibly when dcache is too small. Go for:
762 /usr/vice/etc/afsd -nosettime -stat 12384 -chunk 19
764 > So I ran the full suite of iozone tests (13), but at a single file
765 > size (128M) and one record size (64K). I set the AFS cache size to > 80000K for both memcache and diskcache.
767 Note that memcache size and diskcache size are different things.
768 In the case of memcache, a fixed number of chunks are allocated
769 in memory, such that numChunks * chunkSize = memCacheSize. In
770 the case of disk cache, there are a lot more chunks, because the
771 disk cache assumes not every chunk will be filled (the underlying
772 filesystem handles disk block allocation for us). Thus, when you
773 have small file segments, they use up an entire chunk worth of
774 cache in the memcache case, but only their size worth of cache
775 in the diskcache cache.
779 ### <a name="3.34 Settting up PAM with AFS"></a> 3.34 Settting up PAM with AFS
783 auth sufficient /lib/security/pam_afs.so debug try_first_pass ignore_root debug
784 auth required /lib/security/pam_env.so
785 auth sufficient /lib/security/pam_unix.so likeauth nullok
786 auth required /lib/security/pam_deny.so
788 account required /lib/security/pam_unix.so
790 password required /lib/security/pam_cracklib.so retry=3 type=
791 password sufficient /lib/security/pam_unix.so nullok use_authtok md5 shadow
792 password required /lib/security/pam_deny.so
794 session sufficient /lib/security/pam_afs.so set_token
795 session required /lib/security/pam_limits.so
796 session required /lib/security/pam_unix.so
798 # reafslog is to unlock dtlogin's screensaver
799 other auth sufficient /usr/athena/lib/pam_krb4.so reafslog
801 ### <a name="3.35 Setting up AFS key in KDC a"></a> 3.35 Setting up AFS key in KDC and [[KeyFile]]
803 The easiest method is not much recommended, as the password is possibly very weak:
805 dattan# bos addkey server -kvno <n+1>
809 dattan# kstring2key -c <cellname>
812 dattan# kadmin ckey afs
813 <cut-n-paste the ascii output from above>
815 Beware, there is currently a byte order bug in the kadmin tool as of May 2001 in KTH-KRB. If you run kadmin on a little indian machine (PC, Alpha, ...) you must swap the bytes manually. Here is a small example:
817 bg$ kstring2key -c sics.se
819 ascii = \361\334\211\221\221\206\325\334
820 hex = f1dc89919186d5dc
823 bg.admin@SICS.SE's Password:
824 New DES key for afs: \221\211\334\361\334\325\206\221
826 Please note how the bytes are swapped in groups of four.
828 This bug is fixed in an upcoming release.
830 Better approach is to create random key in KDC using "ksrvutil get" (rather then "ksrvutil add" which asks you for the new password), export it into /etc/srvtab ("ksrvutil ext") and from there copy it into /usr/afs/etc/KeyFile using asetkey utility from /afs/transarc.com/public/afs-contrib. After some years, you may wish to regenerate the random afs key using "ksrvutil change" in KDC and exporting it again.
832 There are two nice pages abou this:
834 <http://www.cs.cmu.edu/afs/andrew.cmu.edu/usr/shadow/www/afs/afs-with-kerberos.html> <http://www.contrib.andrew.cmu.edu/~shadow/afs/afs-with-kerberos.html>
836 As someone noted, you might need some ext\_srvtab utility to extract the key from /etc/srvtab, but I don't remeber that I needed it. Then he also suggets to use asetkey utility.
838 Create afs.natur\\.cuni\\.cz@NATUR.CUNI.CZ principal entry in kerberos:
841 principal instance: natur.cuni.cz
842 principal realm: NATUR.CUNI.CZ
844 Import this key from Kerberos (/etc/srvtab) into AFS /usr/afs/etc/KeyFile using asetkey utility, which is I think from /afs/transarc.com/public/afs-contrib
846 ./asetkey add 0 /etc/srvtab <afs.natur.cuni.cz@NATUR.CUNI.CZ>
848 This [[KeyFile]] with the AFS-key you can just always re-copy to new AFS **server** machines. Be sure that the key version number KVNO is always same in this [[KeyFile]] and in Kerberos database. On all these machines you of course need afs.hostname.REALM key loaded into their /etc/srvtab (create them with 'ksrvutil get'). It seems klog(1) needs afs@REALM, so copy the principal afs.cell.name@REALM to it.
850 You can test if you have same keys in kerberos and AFS like this:
855 If you have some tokens now, then it works and you can now shutdown kaserver. Users, which you have created in AFS under kaserver are in /usr/afs/db/kaserver.\*, but you can just forget them. Create these users again in Kerberos IV.
857 AFS users are then looked up in:
859 kerberos database (/var/kerberos/principal.db)
861 /usr/afs/etc/UserList (permission to manipulate 'bos')
863 system:administrators (permissions for vos, pts, fs, i.e. 'pts adduser' etc)
865 ---------------------------------------------------------------------------------
866 Another approach, untested:
868 http://www.natur.cuni.cz/~majordomo/krb4/krb4.9703/msg00002.html
870 Date: Tue, 9 May 2000 10:57:23 +0200 (MET DST)
871 From: Johan Hedin <johanh at fusion.kth.se>
872 To: Erland Fristedt <Erland.Fristedt@wcs.eraj.ericsson.se>
874 Subject: Re: AFS+KTH-KRB problem
876 This is a printout how we got it to work. Hope this helps.
880 callisto#/usr/athena/sbin/ksrvutil -p johanh.admin -f srvtab get
882 Instance [callisto]: fusion.kth.se
883 Realm [FUSION.KTH.SE]:
884 Is this correct? (y,n) [y]
885 Add more keys? (y,n) [n]
886 Password for johanh.admin@FUSION.KTH.SE:
887 Warning: Are you sure `fusion.kth.se' should not be `fusion'?
888 Added afs.fusion.kth.se@FUSION.KTH.SE
889 Old keyfile in srvtab.old.
890 callisto#/usr/athena/sbin/ksrvutil -p johanh.admin -f srvtab list
892 4 afs.fusion.kth.se@FUSION.KTH.SE
893 callisto#/tmp/asetkey add 4 srvtab afs.fusion.kth.se@FUSION.KTH.SE
895 callisto#bin/bos status callisto
896 Instance buserver, currently running normally.
897 Instance ptserver, currently running normally.
898 Instance vlserver, currently running normally.
899 Instance fs, currently running normally.
900 Auxiliary status is: file server running.
901 callisto#bin/fs la /afs
902 Access list for /afs is
905 system:administrators rlidwka
907 --------------------------------------------------------------------------------
908 > Tokens held by the Cache Manager:
910 > User's (AFS ID 6020) tokens for afs@sunrise.ericsson.se [Expires May 5
914 You got your tickets alright, but then
916 > > touch /afs/.sunrise.ericsson.se/tabort
917 > afs: Tokens for user of AFS id 6020 for cell sunrise.ericsson.se are
918 > discarded (rxkad error=19270407)
920 grep 19270407 /usr/afsws/include/rx/*
921 /usr/afsws/include/rx/rxkad.h:#define RXKADBADTICKET (19270407L)
923 so I would guess that the key version number and/or the key on your
924 AFS servers does not match the key of the afs principal.
926 Use klist -v to figure out about key version numbers and then use
928 $ kstring2key -c sunrise.ericsson.se
930 ascii = \242\345\345\260\323p\263\233
931 hex = a2e5e5b0d370b39b
933 or some other means to ensure that you use the same keys.
935 ### <a name="3.36 Obtaining and getting asetk"></a> 3.36 Obtaining and getting asetkey compiled
937 Better approach is to create random key in KDC, export it into /etc/srvtab and from there move in /usr/afs/etc/KeyFile using asetkey utility from /afs/transarc.com/public/afs-contrib
939 | I can not get asetkey to compile.
941 | asetkey.c: In function `main':
942 | asetkey.c:25: `AFSCONF_SERVERNAME' undeclared (first use in this function)
943 | asetkey.c:25: (Each undeclared identifier is reported only once
944 | asetkey.c:25: for each function it appears in.)
947 AFS 3.5 got rid of a few commonly-used AFS defines; I suspect they are now API calls of some kind, akin to the POSIX move from hardcoded constants to use of \{,f\}pathconf() and sysconf(). I have no idea what those calls are, though.
949 In the meantime, the following two defines are useful in building stuff on AFS 3.5:
951 #define AFSCONF_CLIENTNAME "/usr/vice/etc"
952 #define AFSCONF_SERVERNAME "/usr/afs/etc"
954 ### <a name="3.37 afs_krb_get_lrealm() using"></a><a name="3.37 afs_krb_get_lrealm() using "></a> 3.37 afs\_krb\_get\_lrealm() using /usr/afs/etc/krb.conf
956 In this file you can set also another REALM to be used by you afs server processes, if the REALM should differ from the system-wide REALM (
962 Don't forget it's related to these entries in Kerberos KDC:
965 krbtgt CELL.NAME@REALM
967 If you use heimdal (or MIT krb5), do:
969 echo "REALM" > /usr/afs/etc/krb.conf
971 ### <a name="3.38 Moving from kaserver to Hei"></a> 3.38 Moving from kaserver to Heimdal KDC
973 > While converting all our administration tools, we have discovered that
974 > the time a principal changed his/her/its password is _not_ carried over
975 > from the kaserver DB.
977 First of all, some Heimdal's configure flags:
981 requires krb4 libs, so for that you'll need a working krb4 are you still using a kaserver/kaserver emulation ?
985 is just for dumping a kaserver krb4 database. If you are no longer running a kaserver, you don't need it.
989 <https://lists.openafs.org/pipermail/openafs-info/2002-May/004326.html>
991 This works while migrating from kaserver:
993 /usr/heimdal/libexec/hprop --source=kaserver --cell=xxx
994 --kaspecials --stdout | /usr/heimdal/libexec/hpropd --no-inetd --stdin
996 This somewhat doesn't:
998 /usr/heimdal/libexec/hprop --source=kaserver --cell=xxx
999 --encrypt --master-key=<path to key> --kaspecials --stdout |
1000 /usr/heimdal/libexec/hpropd --stdin
1002 ### <a name="3.39 Moving from KTH-KRB4 to Hei"></a> 3.39 Moving from KTH-KRB4 to Heimdal KDC
1004 /usr/heimdal/libexec/hprop -n -k /etc/krb5.keytab --source=krb4-dump -d /var/kerberos/dump.txt --master-key=/.k -D | /usr/heimdal/libexec/hpropd -n
1008 1. dump of the krb4 database with kdb_util
1009 2. dump of the "default" heimdal database with kadmin -l
1010 3. /usr/heimdal/libexec/hprop -n -k /etc/krb5.keytab --source=krb4-dump -d /usr/heimdal/libexec/hprop -n -k /etc/krb5.keytab --source=krb4-dump -d
1011 where /etc/krb5.keytab contains hprop/`hostname` keys
1012 4. merge of the converted database with file from (2) via kadmin
1014 The special thing for me is the use of "-D" in the (3) which seems to
1015 cause conversion des-cbc-sha1 keys of old krb4 database entries to
1018 ### <a name="3.40 What are those bos(1) -type"></a> 3.40 What are those bos(1) -type values simple and cron?
1020 Typically, admins do this once they configure new afs server:
1022 bos create foo ptserver simple /usr/afs/bin/ptserver -cell bar.baz
1023 bos create foo vlserver simple /usr/afs/bin/vlserver -cell bar.baz
1024 bos create foo fs fs /usr/afs/bin/fileserver /usr/afs/bin/volserver /usr/afs/bin/salvager -cell bar.baz
1026 Type "simple" has one process. type "cron" gets forked at the appropriate time.
1028 ### <a name="3.41 KDC listens on port 88 inst"></a> 3.41 KDC listens on port 88 instead of 750
1030 "Kramer, Matthew" <mattkr@bu.edu> writes:
1033 > Our Kerberos 4 environment listens on only port 750 and doesn't
1034 > listen on port 88. For Win2K/XP machines this causes a problem in
1035 > C:\openafs-1.2.8\src\kauth\user_nt.c line 144 when the Kerberos port is
1036 > queried from the %systemroot%\system32\drivers\etc\services file. Win2K
1037 > and XP both return 88 which is correct for Kerberos 5 but in our case
1038 > not for Kerberos 4. Why doesn't it instead query for kerberos-iv which
1039 > would return the correct value of 750?
1041 > Y:\src\kauth>diff user_nt.c.orig user_nt.c
1043 > < sp = getservbyname("kerberos", "udp");
1045 > > sp = getservbyname("kerberos-iv", "udp");
1047 A quick work around could be to use Loves requestforwarder found at
1049 ftp://ftp.stacken.kth.se/pub/projects/krb-forward/krb-forward-0.1.tar.gz
1053 ### <a name="3.42 afsd gives me "Error -1 in"></a><a name="3.42 afsd gives me "Error -1 in "></a> 3.42 afsd gives me "Error -1 in basic initialization." on startup
1055 When starting afsd, I get the following:
1057 # /usr/vice/etc/afsd -nosettime -debug
1058 afsd: My home cell is 'foo.bar.baz'
1059 ParseCacheInfoFile: Opening cache info file '/usr/vice/etc/cacheinfo'...
1060 ParseCacheInfoFile: Cache info file successfully parsed:
1061 cacheMountDir: '/afs'
1062 cacheBaseDir: '/usr/vice/cache'
1064 afsd: 5000 inode_for_V entries at 0x8075078, 20000 bytes
1065 SScall(137, 28, 17)=-1 afsd: Forking rx listener daemon.
1066 afsd: Forking rx callback listener.
1067 afsd: Forking rxevent daemon.
1068 SScall(137, 28, 48)=-1 SScall(137, 28, 0)=-1 SScall(137, 28, 36)=-1 afsd: Error -1 in basic initialization.
1070 Solution: make sure you have kernel module loaded.
1072 ### <a name="3.43 Although I get krb tickets,"></a> 3.43 Although I get krb tickets, afslog doesn't give me tokens, I see UDP packets to port 4444
1074 Using tcpdump I see the following traffic. What runs on the port 4444?
1076 15:16:54.815194 IP foo.bar.baz.32772 > foo.bar.baz.4444: UDP, length: 269
1077 15:16:54.815199 IP foo.bar.baz > foo.bar.baz: icmp 305: foo.bar.baz udp port 4444 unreachable
1079 This is the 5->4 ticket conversion service on UDP port 4444. Assuming your afs servers know how to deal with krb5 tickets (since openafs-1.2.10? and 1.3.6x?) include the following in /etc/krb5.conf:
1084 ### <a name="3.44 I get error message trhough"></a> 3.44 I get error message trhough syslogd: "afs: Tokens for user of AFS id 0 for cell foo.bar.baz are discarded (rxkad error=19270407)"
1088 grep 19270407 /usr/afsws/include/rx/*
1089 /usr/afsws/include/rx/rxkad.h:#define RXKADBADTICKET (19270407L)
1091 Note that a common cause of this problem is the use of periods (".") in kerberos principals. If you have a kerberos principal such as my.name@REALM.COM and create the corresponding pts userid "my.name" you will get the cryptic error above. If you want to use such principal names and have OpenAFS 1.4.7 or later, you can pass the option -allow-dotted-principals to all server daemons to allow their use. See the -allow-dotted-principals option in the fileserver (or any server daemon) documentation for more information.
1093 ### <a name="3.45 I get tickets and tokens, b"></a> 3.45 I get tickets and tokens, but still get Permission denied.
1095 Answer: /usr/afs/etc/UserList accepts only krb4 syntax. Use `joe.admin` instead of `joe/admin`. See `https://lists.openafs.org/pipermail/openafs-devel/2002-December/008673.html` and the rest of the thread.
1097 ### <a name="3.46 Recovering broken afs cache"></a> 3.46 Recovering broken afs cache on clients
1099 >> Does anyone have a trick to force AFS to refresh its cache (for a
1100 >> particular directory or even for all files?) The only way I know
1101 >> how to accomplish this is to reboot, stop in single user mode,
1102 >> rm -rf the cache files and let AFS rebuild everything.
1104 > fs flush and fs flushv have cured corruption problems in the past
1105 > on some of our clients.
1107 Thanks for the tip - I was not aware of the flush* subcommands.
1108 Here's a little of what I saw today:
1111 /bin/ls: asso.S14Q00246.all.log: Bad address
1112 /bin/ls: asso.S14Q00246.all.lst: Bad address
1113 /bin/ls: chr14markers.txt: Bad address
1114 /bin/ls: geno.summary.txt: Bad address
1115 /bin/ls: global.ind.S14Q00246.all.txt: No such device
1116 /bin/ls: global.S14Q00246.all.txt: No such device
1118 [ other ls results as usual ]
1120 Flushing a particular file had no effect (the same error as shown above appears). Flushvolume took a long time, but when it eventually completed, the ls -la behaved exactly as one would expect.
1122 ### <a name="3.47 What does it mean for a vol"></a> 3.47 What does it mean for a volume to not be in the VLDB?
1124 If a volume is not in the VLDB, you will be unable to perform operations on it using its name; all "vos" operations will need to be done using its numerical id, server, and partition. Furthermore, if a volume is not in the VLDB, it cannot be reached via mountpoints.
1126 ### <a name="3.48 What is a Volume Group?"></a> 3.48 What is a Volume Group?
1128 You can think of a Volume Group as an RW volume, and all of the clones of that RW (its RO clone, BK clone, and any other clones). All of the volumes in a Volume Group can share storage for data that is the same between all of them. This is why, for example, an RO clone usually takes up very little disk space; since an RW and its RO clone are in the same Volume Group, they can share storage for unchanged data.
1130 All of the volumes in a group usually have very similar volume ID numbers. For example, if an RW volume has ID 536870915, its RO clone will typically be 536870916. However, this is not required, as volume ID numbers can be almost anything. You can even manually specify what volume ID number you want for a volume when you create the volume with "vos create".
1132 Currently, you can only have about 8 volumes in a Volume Group. However, this limitation is due to technical details of the fileserver "namei" disk backend. If that backend is improved in the future, or if different backends are developed, the number of volumes in a volume group could be much greater.
1134 Some low-level documentation may refer to a "volume group ID". This is always the same as the RW volume ID.
1136 ### <a name="3.49 What is a Clone?"></a> 3.49 What is a Clone?
1138 A clone of a volume is a read-only copy of that volume which shares on-disk storage with the original volume. Backup volumes are a particular kind of clone volume. Read-only replicas which reside on the same partition as their read-write volume are another particular kind of clone volume. In some other storage systems this kind of volume is called a "snapshot".
1140 Clone volumes must belong to the same volume group (see previous question) as the volume which they are a clone of.
1142 In addition to backup and readonly clones, you may create up to three additional clones of a volume. To do this, use "vos clone".
1144 When you "vos remove" a volume, its "backup" clone will also be removed automatically. However, clones created with "vos clone" are **not** removed automatically. Unfortunately, these "dangling clones" will no longer be in the VLDB (see question 3.47). They belong to a volume group whose leader (rw volume) no longer exists, which is a somewhat undefined state for AFS. Such volumes should be manually deleted as soon as possible.
1146 ### <a name="3.50 What is a Shadow?"></a> 3.50 What is a Shadow?
1148 A shadow of a volume is a duplicate of that volume which resides on a different partition. Shadow volumes do not share storage with their original volumes (unlike clones). A readonly volume on a **different** partition than its readwrite volume could be considered one particular example of a shadow volume; however, in practice the term "shadow volume" is used to refer to volumes created with "vos shadow" and not to refer to readonly volumes.
1150 A shadow of any readwrite volume may be created using the "vos shadow" command. This will create a new volume which is a shadow of the original volume, and will copy the contents of the old volume to the new volume. This will also set a bit in the header of the new shadow volume that identifies it as a shadow volume. Shadow volumes do not appear in the VLDB (see question 3.47) -- "vos shadow" does not create a VLDB entry and "vos syncvldb" ignores shadow volumes.
1152 You can "refresh" a shadow volume from its original with "vos shadow -incremental". This operation will first check to make sure that the target of the operation is a shadow volume, to prevent the administrator from accidentally corrupting a non-shadow volume. However, if you shadow from a readwrite volume to some shadow of **another** volume, the shadow will be corrupted and will have to be deleted. Vos shadow will only copy data which has changed, so it is very efficient.
1154 You can remove the shadow bit from a volume's header with "vos syncvldb -force". This will remove the shadow bit and create a VLDB entry for the volume, deleting any previous entry for the rw volume. However, the rw volume itself will not be deleted; it will simply exist without a VLDB entry.
1156 Attempting to create shadows of two different rw volumes on the same partition with the same name is prohibited by the volserver. Technically it is possible to create two shadow volumes with the same name on different partitions; however, this is not advisable.
1158 ### <a name="3.51 Can I authenticate to my af"></a> 3.51 Can I authenticate to my afs cell using multiple kerberos Realms?
1160 Yes. This can be useful if your organization has multiple kerberos Realms with identical user entries: For example you might have an MIT Kerberos realm for Unix-like systems, and an Active Directory domain for windows with synchronized accounts.
1162 As long as you only need to support 2 realms, and one of them has the same name as the afs cell(or the uppercase of the name of the afs cell) you can do this without upgrading to a new version of AFS. If you need more Realms, or if neither Realm name matches your cell name, there is work to support this in the development tree.
1164 In order to make this work, you need to do 4 things.
1166 1) add a key for the afs service to the additional realm and store it in a keytab:
1168 $ kadmin -q ank -e des-cbc-crc:v4 -kvno <new kvno> afs/your.cell.name@YOUR.SECOND.REALM.NAME
1169 $ kadmin -q ktadd -e des-cbc-crc:v4 -k /path/to/afs.keytab afs/your.cell.name@YOUR.SECOND.REALM.NAME
1171 Note that a kvno must be specified for the key that is different than the kvno for your existing key(s) in the original realm. you can check on the kvno of the existing keys by running "asetkey list" on one of your servers. since keys must be in ascending order in the AFS [[KeyFile]] it will be easiest if you make the new kvno higher than any existing key's kvno.
1173 It's also worth noting that the process of adding the key to a keytab(at least with MIT krb5) actually creates a new key first, so your kvno will end up being higher than what you specified when you added the principal. you can check on the current kvno by using the "kadmin -q getprinc afs/your.cell.name@YOUR.SECOND.REALM.NAME" command.
1175 2) add the new key to the afs [[KeyFile]] on your afs servers:
1177 $ asetkey add <kvno> /path/to/afs.keytab afs/your.cell.name@YOUR.SECOND.REALM.NAME
1179 note that the kvno here needs to be the same one as is reported by the kadmin getprinc command.
1181 3) create an afs krb.conf with your additional Realm's name in it, and place it on all of your AFS servers:
1183 echo "YOUR.SECOND.REALM.NAME" > /usr/afs/etc/krb.conf
1185 4) restart you afs servers.
1187 at this point you should be able to run:
1189 kinit you@YOUR.SECOND.REALM.NAME
1192 and recieve the same privileges as if you had run:
1194 kinit you@YOUR.CELL.NAME
1197 ### <a name="3.52 Is it a good idea to store"></a> 3.52 Is it a good idea to store mail in AFS?
1199 Some people have expressed the opinion that this is not a good idea. Assuming that a modern (one-file-per-message) format is being used, the concrete reasons to avoid storing mail in AFS include:
1201 1. AFS is not designed to provide good performance under the sort of concurrent accesses that occur when many mail server machines share a common space in AFS; see [this message](http://www.openafs.org/pipermail/openafs-info/2007-June/026621.html) for more information.
1202 - This does not apply to situations where there is a single mail server for a particular user.
1203 - This also may not apply if [Callback Extended Information](http://michigan-openafs-lists.central.org/archives/afs3-standardization/2008-May/000123.html) is implemented.
1205 ### <a name="3.53 How can I ensure that the"></a><a name="3.53 How can I ensure that the "></a> 3.53 How can I ensure that the userids on client machines match the users' pts ids?
1207 You can use [libnss-afs](http://www.hcoop.net/~megacz/software/libnss-afs.html) for this.
1209 ### <a name="3.54 What is Fast Restart?"></a> 3.54 What is Fast Restart?
1211 When compiled with --enable-fast-restart, the file server will start up immediately, without first salvaging any volumes which cannot be attached.
1213 Disadvantages to Fast Restart, [as noted here](http://lists.openafs.org/pipermail/openafs-info/2008-May/029386.html) include:
1215 1. Volumes in need of salvage remain offline until an administrator intervenes manually
1216 2. On an inode-based fileserver, salvaging a single volume crawls every inode; therefore, salvaging volumes individually (rather than partition-at-a-time) duplicates work.
1218 ### <a name="#3.55 Reboot ?"></a> 3.55 Why does AFS reboot itself spontaneously at 4:00am every Sunday?
1220 This was made to be the default behavior back in the days when OpenAFS servers had problems with leaking memory and other resources. These days, it is generally seen as not necessary.
1222 You can disable this behavior with:
1224 bos setrestart $SERVER_NAME never
1225 bos setrestart $SERVER_NAME -newbinary never
1227 ### <a name="#3.56 weak crypto"></a>3.56 Why do I get an error -1765328370 when authenticating?
1229 -1765328370 KDC has no support for encryption type
1231 In other words, your KDC has support for DES-CBC-CRC turned off.
1232 Re-enable support for DES encryption types and you will get further.
1234 Check /etc/krb5.conf and make sure it has "allow_weak_crypto = true" in the [libdefaults]
1235 section. Also check kdc.conf (possibly located in /var/kerberos/krb5kdc) and make sure
1236 that des-cbc-crc:normal is in the supported_enctypes list.