Dave Tanner [Sun, 25 Jun 2006 20:50:37 +0000]
windows-torture-test-
20060625
Torture Test
Jeffrey Altman [Sun, 25 Jun 2006 20:50:02 +0000]
windows-largefile-test-
20060625
Large File test
Qi Dong [Sun, 25 Jun 2006 18:56:53 +0000]
windows-aklog-64bit-
20060625
Prepare makefile to link against kfw 64-bit libs when they are
available.
Qi Dong [Sun, 25 Jun 2006 18:54:16 +0000]
windows-64bit-type-safety-
20060625
Type casting
Jeffrey Altman [Sun, 25 Jun 2006 17:55:23 +0000]
windows-cleanup-
20060625
cleanup AFS_component_version_number.h
Jeffrey Altman [Sun, 25 Jun 2006 15:59:38 +0000]
windows-wix-tools-on-i386-only-
20060625
only build the 32-bit tools installer on i386
Jeffrey Altman [Sun, 25 Jun 2006 15:54:25 +0000]
windows-nsis-help-files-install-location-
20060625
place the help files in the correct location
Jeffrey Altman [Sun, 25 Jun 2006 02:47:04 +0000]
windows-report-quota-diskfull-
20060624
overquota and diskfull errors are detected while writing dirty buffers
to the file server. A flag is set on the scp so that the error can
be returned to the caller. cm_FSync was clearing the flag without
returning the error. This permitted calls to Win32 WriteFile to report
success even when they failed.
Russ Allbery [Sat, 24 Jun 2006 17:08:40 +0000]
rhel3-krb524-checking-
20060624
FIXES 30589
Check for krb524 functions in libkrb524 if not found in libkrb5 so that the
Kerberos v5 support will build on RHEL3. Thanks to Derek Atkins for the
debugging and testing.
Jeffrey Altman [Sat, 24 Jun 2006 15:09:34 +0000]
windows-viced-cbd-
20060624
don't break platforms other than Windows
Christopher Allen Wing [Sat, 24 Jun 2006 06:47:09 +0000]
afsutil-includes-
20060623
Include <afs/afsutil.h> for prototypes of afs_inet_ntoa and afs_inet_ntoa_r.
Russ Allbery [Sat, 24 Jun 2006 06:14:09 +0000]
autoconf-datarootdir-
20060623
Autoconf 2.60 will replace some variables, most notably mandir, with values
relative to datarootdir, a new variable. Add settings of datarootdir to
the affected files.
Asanka Herath [Sat, 24 Jun 2006 03:48:06 +0000]
windows-wix-help-files-install-location-
20060623
place the help files in the correct location
Asanka Herath [Fri, 23 Jun 2006 08:54:27 +0000]
windows-largefile-support-
20060623
This patch adds large file support (> 2GB) to the CIFS server and
the cache manager.
It also corrects a significant oversight where the definition of
struct rx_call was inconsistent due to a failure to define AFS_PTHREAD_ENV
prior to including rx.h in all source files. This prevented all errors
during rx_Read and rx_Write to be ignored.
Jeffrey Altman [Fri, 23 Jun 2006 04:32:48 +0000]
windows-cm_syncop-buflocked-
20060622
If cm_SyncOp is called with the buffer locked, it must be returned
locked even on error.
Jeffrey Altman [Thu, 22 Jun 2006 22:48:32 +0000]
windows-no-freelance-vldb-failover-
20060622
When Freelance mode is not in use the cache manager must load the root.afs
volume of the default cell in order to be able to start. Unfortunately,
if the first vlserver fails the request to obtain the volume information
will timeout and no retries will be performed. This patch performs up
to 10 retries in the initialization code in order to enable the vlservers
to be marked down so that failover can occur.
Jeffrey Altman [Wed, 21 Jun 2006 21:40:00 +0000]
windows-viced-cbd-
20060621
typo
Jeffrey Altman [Tue, 20 Jun 2006 23:09:46 +0000]
windows-viced-cbd-
20060620
build the callback debug utility
Tom Keiser [Wed, 21 Jun 2006 18:57:24 +0000]
aix-largefile-cleanup-
20060620
some largefile cleanup and other minor aix tweaking
Jeffrey Altman [Tue, 20 Jun 2006 21:57:26 +0000]
windows-afsrpc-exports-
20060620
New functions: rx_InitHost, rx_NewServiceHost
Jeffrey Altman [Tue, 20 Jun 2006 21:32:55 +0000]
rxbind-for-servers-
20060620
do not call parseNetFile on Windows where the function does not exist
Derrick Brashear [Tue, 20 Jun 2006 21:07:06 +0000]
revert-rx-magic-check-
20060607
revert until i recode to avoid conflict with queues
Jeffrey Hutzelman [Tue, 20 Jun 2006 21:00:09 +0000]
viced-findclient-do-not-return-wrong-client-
20060620
In h_FindClient_r it was possible for the wrong client to be returned
and worse for it to be returned without the reference count being
incremented.
Rainer Toebbicke [Tue, 20 Jun 2006 15:50:17 +0000]
viced-deadlock-
20060620
FIXES 34073
* checks for client == oldClient
* drops the H_LOCK prior to obtaining the WriteLock on oldClient
in order to prevent a deadlock
Jim Rees [Fri, 16 Jun 2006 20:08:29 +0000]
no-divide-by-zero-
20060616
Print nothing rather than dividing by zero if the cache manager doesn't
return excessive cache params.
Troy Benjegerdes [Thu, 15 Jun 2006 16:51:25 +0000]
fs-getcache-show-more-
20060615
show more info about cache usage, optionally
Troy Benjegerdes [Thu, 15 Jun 2006 16:42:24 +0000]
linux-current-already-used-
20060615
don;'t conflict on use of current
Derrick Brashear [Thu, 15 Jun 2006 16:12:13 +0000]
getaddr-avoid-looping-
20060615
so it would be good to count these all off
oops
Derrick Brashear [Tue, 13 Jun 2006 01:28:48 +0000]
rxbind-for-servers-
20060612
FIXES 32308
support -rxbind switch for servers
Derrick Brashear [Mon, 12 Jun 2006 17:19:27 +0000]
linux-enable-gcpags-
20060612
enable gcpags
Chas Williams [Mon, 12 Jun 2006 05:23:54 +0000]
recvackpkt-count-acks-and-open-cong-window-
20060612
"the non AFS_GLOBAL_RXLOCK_KERNEL version of
rxi_ReceiveAckPacket() wasnt counting 'hard' acks and not opening the
congestion window. using rx_Readv() directs the incoming packets into
the target iov, and sends a hard ack (skipping soft acks). "
Derrick Brashear [Tue, 13 Jun 2006 04:50:03 +0000]
getaddr-skip-loopbacks-version-for-macos-
20060613
so rxbind does something sensible
Derrick Brashear [Tue, 13 Jun 2006 00:58:33 +0000]
namei-dec-drop-lock-
20060612
FIXES 17638
unsure why yet, or even if, but this is all that makes sense
Jeffrey Altman [Thu, 8 Jun 2006 22:17:04 +0000]
windows-scache-recycle-fix-
20060608
do not special case freelance entries
Jeffrey Altman [Thu, 8 Jun 2006 00:28:40 +0000]
windows-cifs-std-info-take-two-
20060607
The "is a directory" value was being set one field beyond the end of
the Standard Info reply structure.
Chaskiel M Grundman [Wed, 7 Jun 2006 05:56:55 +0000]
no-client-tcon-
20060607
remove client->tcon and avoid locking issues
Chaskiel M Grundman [Wed, 7 Jun 2006 05:31:33 +0000]
rx-magic-check-
20060607
for pullup as /afs/andrew/usr18/cg2v/rx-magic-check.diff
Chaskiel M Grundman [Wed, 7 Jun 2006 05:25:38 +0000]
openlog-consistency-
20060607
open logs consistently so assert logging is appended
Russ Allbery [Tue, 6 Jun 2006 23:33:07 +0000]
synchronize-xdr-
20060606
Include the same set of xdr objects in the shared libafsrpc as are included
in the static libafsrpc.
Jeffrey Altman [Tue, 6 Jun 2006 17:57:43 +0000]
windows-smb-mountpt-is-dir-
20060606
Be consistent about reporting mountpoints and invalid links as directories
to the CIFS client.
Jeffrey Altman [Tue, 6 Jun 2006 15:43:16 +0000]
windows-makefile-clean-
20060606
cleanup additional files
Jeffrey Altman [Mon, 5 Jun 2006 04:25:03 +0000]
windows-version-
20060604
1.5.0002
Jeffrey Altman [Mon, 5 Jun 2006 02:29:19 +0000]
windows-notes-
20060604
notes
Jeffrey Altman [Sun, 4 Jun 2006 07:11:53 +0000]
viced-swap-old-cbcon-diff-
20060603
FIXES 32985
swap cbconn when other side changes IPs
Jeffrey Altman [Sat, 3 Jun 2006 18:59:35 +0000]
windows-scache-recycle-
20060603
A deadlock was detected when performing "fs flushall" if the file
server reports VNOVNODE. The scp->createBufferLock is already
held by the current thread and there is no mechanism to propagate
the knowledge. Therefore, an alternate mechanism for clearing the
cache must be developed. A new function cm_RecycleSCache(scp,flags)
has been extracted from cm_GetNewSCache(). This function performs
the task of recycling an cm_scache_t object. When called from
cm_GetNewSCache() with no flags the expectation is that there are
no associated buffers that are queued to be read or written. When
called from cm_Analyze() with the CM_SCACHE_RECYCLEFLAG_DESTROY_BUFFERS
flag, any queued buffers will be de-queued and marked as if the operations
were performed so the data can be discarded.
This patch also addresses the following issues:
* CM_ERROR codes were logged as "unknown" by cm_Analyze.
* In response to VNOVNODE, the parent is only discarded if the current
cm_scache_t is not a directory.
* In the Ioctl FlushFile and FlushVolume functions, there were no
checks to protect against flushing the Freelance SCache entries.
* In FlushFile, the wrong cm_scache_t object was being released.
* In cm_GetNewSCache, do not allow Freelance SCache entries to be
recycled. Choose a new entry if cm_RecycleSCache fails.
Chaskiel M Grundman [Fri, 2 Jun 2006 22:12:27 +0000]
flock-handle-uae-errors-
20060601
handle uae versions of errors for flock locks
Joe Buehler [Thu, 1 Jun 2006 16:57:15 +0000]
linux-vfsops-remove-redundant-code-
20060601
FIXES 32783
remove redundancy
Jeffrey Altman [Thu, 1 Jun 2006 04:50:13 +0000]
windows-smb-fid-faster-
20060531
Speed up the performance of the cache manager by not holding the
smb_fid_t mutex across calls to cm_SyncOp and AFS RPCs.
Ensure that all smb_fid_t flag references are protected.
Derrick Brashear [Wed, 31 May 2006 21:38:30 +0000]
Jeffrey Altman [Tue, 30 May 2006 17:56:51 +0000]
windows-smb-fid-deadlock-
20060530
remove a deadlock introduced in 1.4.1
Jeffrey Altman [Mon, 29 May 2006 00:03:52 +0000]
windows-get-acls-
20060528
It is possible to have a valid callback but not to have the
required ACL info for the current user. Force acquisition of
a FetchStatus but do so without making multiple calls.
Jeffrey Altman [Mon, 29 May 2006 00:01:26 +0000]
windows-hexdump-fix-
20060528
do not allow unquoted percent signs to be logged
Jeffrey Altman [Fri, 26 May 2006 22:11:15 +0000]
windows-access-fetchstatus-
20060526
Do not call cm_GetCallback() directly. Always
use cm_SyncOp(CM_SCACHESYNC_NEEDCALLBACK)
Jeffrey Altman [Fri, 26 May 2006 07:35:34 +0000]
windows-dynamic-thread-priority-
20060525
Dynamically adjust the priority of server threads based upon the age
of the cifs request that is being processed. Bump the priority one
level for each 15 seconds of age.
Jeffrey Altman [Thu, 25 May 2006 18:11:57 +0000]
windows-head-tail-queue-removal-
20060525
while investigating the cause of the delayed write errors it was observed
that all of the sleep queues are LIFO. This has the side effect of
encouraging starvation. Changing the queues to FIFOs revealed a serious
problem affecting the use of all queues which use both head and tail
pointers. The removal function osi_QRemove does not take a tail pointer
and therefore the pointer is always left hanging. If the number of elements
ever drops to zero the queue becomes corrupted.
Added osi_QRemoveHT to be used whenever head and tail pointers are used.
Updated all callers in afsd.
Jeffrey Altman [Wed, 24 May 2006 23:59:51 +0000]
windows-smb-mods-
20060524
Prevent smb_FindVCP from returning dead virtual circuits
Treat "*." as an alias for "all"
Derrick Brashear [Wed, 24 May 2006 16:49:11 +0000]
getaddr-skip-loopbacks-version-for-macos-
20060523
so rxbind does something sensible
Hartmut Reuter [Tue, 23 May 2006 17:29:11 +0000]
rxkad-var-tkt-len-
20060523
FIXES 31966
allocate memory for struct rxkad_cprivate based upon the actual
ticket size not the max ticket size
Jeffrey Altman [Tue, 23 May 2006 05:47:29 +0000]
rx-rdwr-type-casting-
20060523
type cast correctly
Jeffrey Altman [Sat, 20 May 2006 04:14:07 +0000]
windows-aklog-afsid-lookup-
20060519
if the cell matches the realm, do not lookup with realm
Jeffrey Altman [Sat, 13 May 2006 15:22:20 +0000]
viced-host-mobile-client-
20060505
Do not delete an old host entry because the UUID because the UUIDs
of the clients at the old addresses are wrong. Remove the addresses.
Do not allow the same host,addr,port combination to be added into
the hash table multiple times.
====================
This delta was composed from multiple commits as part of the CVS->Git migration.
The checkin message with each commit was inconsistent.
The following are the additional commit messages.
====================
Add some more logging
Do not allow 'code' to be re-used for probing the uuid of the old host
primary address because we are relying on the value of 'code' to indicate
whether or not RCallBackConnectBack failed. If it did, we mark the
client as VENUSDOWN, if not we marked it as RESETDONE.
====================
log the old addr, not the new one
Jeffrey Altman [Thu, 11 May 2006 13:36:06 +0000]
windows-smb-vcp-cleanup-races-
20060511
Only allow one thread to mark the vcp dead
Derrick Brashear [Wed, 10 May 2006 05:51:41 +0000]
viced-writevalloc-dont-vtakeoffline-
20060510
random errors shouldn't take the volume offline
Jim Rees [Tue, 9 May 2006 19:41:47 +0000]
fbsd-saddr-
20060509
Small fixes to get sockaddr_storage changes working on FreeBSD
Jim Rees [Tue, 9 May 2006 16:49:49 +0000]
snprintf-
20060509
Change sprint to snprintf
Jeffrey Altman [Tue, 9 May 2006 06:18:08 +0000]
windows-addr-change-mods-
20060509
When the client changes IP addresses, in order to obtain optimal
performance it must:
(1) force the replacement of all existing rx connections
(2) ping all up and down servers
Jeffrey Altman [Tue, 9 May 2006 05:15:47 +0000]
bringing-rx-into-21st-century-
20060504
rx_InitHost() port is in network byte order
Jeffrey Altman [Tue, 9 May 2006 02:37:30 +0000]
windows-less-power-mgmt-logging-
20060508
comment out some of the power mgmt log entries
reducing the size of afsd_init.log
Jeffrey Altman [Tue, 9 May 2006 02:34:20 +0000]
windows-smb-vcp-clean-
20060508
don't grab the vcp mutex after releasing the vcp
Jeffrey Altman [Tue, 9 May 2006 02:30:37 +0000]
windows-osi-lock-validate-
20060508
perform lock type validation
Derrick Brashear [Mon, 8 May 2006 20:26:38 +0000]
macos-packaging-oops-
20060508
no plugin yet
Jeffrey Altman [Fri, 5 May 2006 19:26:51 +0000]
windows-manifests-
20060505
embed manifests when building tools needed for building OpenAFS
Jeffrey Altman [Fri, 5 May 2006 17:27:21 +0000]
viced-host-logging-
20060505
another place where dotted notation and port numbers were not being used
in log output.
Jeffrey Altman [Fri, 5 May 2006 16:50:21 +0000]
windows-wix-vs8-
20060505
Enable wix installers to build with VS8 compiler on X86
Jim Rees [Fri, 5 May 2006 15:37:28 +0000]
cast-sockaddr-
20060505
Cast sockaddr_in to eliminate warning
Ken Hornstein [Fri, 5 May 2006 06:47:37 +0000]
more-rx-updates-
20060504
Suppress warnings from prototype changes.
Trial code for OpenBSD support.
Fix missing piece from UKERNEL.
====================
This delta was composed from multiple commits as part of the CVS->Git migration.
The checkin message with each commit was inconsistent.
The following are the additional commit messages.
====================
IPv6 support for Windows requires the use of ws2tcpip.h and associated
restructing of include file order.
Ken Hornstein [Thu, 4 May 2006 22:23:17 +0000]
bringing-rx-into-21st-century-
20060504
Large series of changes to replace the use of afs_uint32 for IP address
with struct sockaddr_storage in Rx. May even work with IPv6, but only
tested with IPv4. Significant change to internal APIs; existing external
APIs kept, new ones added that take struct sockaddr_storage arguments.
Jeffrey Altman [Thu, 4 May 2006 05:54:19 +0000]
viced-hcps-logging-
20060504
log whenever a client host does not have an hcps value.
log the host by dotted ip-addr and port number
Tommie Gannert [Sun, 30 Apr 2006 05:58:28 +0000]
windows-ansi-filenames-fix-
20060430
FIXES 31202
Remember to perform an OemToChar() conversion of the file name being
queried in Trans2QueryFileInfo
Tommie Gannert [Wed, 26 Apr 2006 16:22:33 +0000]
vlserver-rxmaxmtu-fix-
20060426
use 'index' as offset to argv
Jeffrey Altman [Thu, 27 Apr 2006 16:49:55 +0000]
windows-deadlock-and-race-removal-
20060427
This patch fixes:
* race conditions around cm_Lock() calls that were not protected
by cm_SyncOp(LOCK) [asanka@secure-endpoints.com]
* deadlocks caused by obtaining smb_fid_t->mx after cm_scache_t->mx
* removes an extra Release smb_fid_t->mx that could result in
releasing a mutex that is not currently held
* changes the log representation of several return codes and fids to
be consistent with other output
Rainer Toebbicke [Wed, 26 Apr 2006 16:43:17 +0000]
namei-delay-fsync-
20060426
FIXES 30632
group fsync calls to speed up operations
Derek Atkins [Wed, 26 Apr 2006 16:40:43 +0000]
os-probe-linux-2-6-9-
20060426
FIXES 30514
make this work on 2.6.9
Christof Hanke [Wed, 26 Apr 2006 16:36:21 +0000]
pam-configure-optmz-
20060426
FIXES 30659
get rid of hardcoded optimize flags
Derrick Brashear [Wed, 26 Apr 2006 16:03:20 +0000]
md5-fix-
20060426
the user struct macros hate us
Jeffrey Altman [Wed, 26 Apr 2006 14:56:31 +0000]
linux-probe-missing-commas-
20060426
FIXES 30977
Add missing commas for non-2.6 x86_64 platforms
Jeffrey Altman [Tue, 25 Apr 2006 07:40:48 +0000]
viced-write-locking-fix-
20060425
According to src/libacl/prs_fs.h the PRSFS_LOCK privilege is only supposed
to control read-locks. The PRSFS_WRITE and PRSFS_INSERT privileges are
supposed to control write-locks.
Jeffrey Altman [Mon, 24 Apr 2006 16:02:48 +0000]
windows-version-
20060424
1.5.1
Jeffrey Altman [Mon, 24 Apr 2006 16:00:50 +0000]
windows-notes-
20060424
notes for 1.5.1
Jeffrey Altman [Mon, 24 Apr 2006 15:25:12 +0000]
windows-pthread-cleanup-
20060424
A number of crash reports have been filed with Microsoft in afspthread.dll.
The crashes are occuring as the processes are terminating. In order to
attempt to address this problem this patch adds support for a DllMain()
entry point that is used to cleanup Thread Local Storage and the various
RX queues. One theory is that processes are loading and unloading DLLs
that are linked to afspthread.dll leaving the pthread data in a very
inconsistent state after each unload. By cleaning up the static data
during the unload, if afspthread is loading again it will be forced to
initialize the data once again.
Jeffrey Altman [Sat, 22 Apr 2006 20:44:28 +0000]
windows-insert-locking-without-afsid-
20060422
This patch disables the lookup of the user's AFSID and instead solves
the problem by tracking which cm_user_t created the file. If the user
is unable to obtain a write lock and is the creator, then we check for
the Insert bit. Otherwise, we deny access.
Jeffrey Altman [Sat, 22 Apr 2006 04:45:47 +0000]
windows-vista-
20060421
Vista has a bug in the Netbios ADAPTER_STATUS output in which the
first four bytes of the MAC address with 127.0.0.1. We must check
for that when searching for the Loopback Adapter.
Jeffrey Altman [Fri, 21 Apr 2006 16:08:17 +0000]
windows-notes-
20060421
updates post 1.5.0
Jeffrey Altman [Fri, 21 Apr 2006 16:03:34 +0000]
windows-rx-init-fallback-
20060421
when the cache manager is being frequently restarted (especially during
debugging) the 7001/udp port is often not freed appropriately. This
prevents the cache manager from running. This patch allows the cache
manager to retry with a random port number.
Jim Rees [Fri, 21 Apr 2006 15:58:38 +0000]
openbsd-packaging-
20060421
Use tar and shell scripts instead of pkg_ for packaging. The new package
system apparently depends on ports and can't be used standalone.
Jeffrey Altman [Fri, 21 Apr 2006 14:52:02 +0000]
windows-wix-
20060421
missing close paren
Asanka Herath [Fri, 21 Apr 2006 05:57:55 +0000]
windows-wix-32bit-tools-
20060420
Add support for build a 32-bit tools package for use on 64-bit Windows.
Asanka Herath [Thu, 20 Apr 2006 19:52:03 +0000]
windows-client-side-locking-
20060420
In order to make the client side locking more usable, if the client
is denied a lock by the server and the user only has read privileges,
then we will allocate a local lock. Local locks are not allocated
for write locks.
There is a bug in the file server locking dating back to at least
AFS 3.1 in which the lock privilege is used for read, write and insert
locks. According to the docs, the lock privilege is only supposed to
control read locks. Write locks on new and existing files are supposed
to be controlled by the Write and Insert privileges. This will be fixed
in the file server by another commit.
Jeffrey Altman [Thu, 20 Apr 2006 18:19:22 +0000]
windows-query-afsid-
20060420
In order to properly handle the local allocation of locks
the cache manager must enforce the locking rules associated
with PRSFS_WRITE and PRSFS_INSERT and PRSFS_LOCK. Insert
affects "new files" which in AFS are defined as any file that
is owned by the user. Therefore, we must know the afsid of the
user so that it can be compared to owner of the file.
This commit includes a query using PR_SNameToId where the name
is the name specified by the entity setting the token. There
does not currently exist a Protection Service RPC to return the
AFSID of the entity that the rx connection was authenticated as.
Such a function will have to be added in order to make this
functionality secure.
====================
This delta was composed from multiple commits as part of the CVS->Git migration.
The checkin message with each commit was inconsistent.
The following are the additional commit messages.
====================
Export functions missing from the UNIX version and the PR_ functions
required by the Windows Cache Manager.
Jeffrey Altman [Mon, 17 Apr 2006 19:59:41 +0000]
rx-dpf-
20060417
fix dpf to print the right thing and not simply fail to compile