IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
Till now nfsx would crash if the export was mounted simply as
<volname> as compared to the traditional method of /<volname>.
Signed-off-by: Shehjar Tikoo <shehjart@gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 944 (vmware unmount crashes nfsx)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=944
When a lookup request comes in for (rootfh, ".."), we need to handle
it in a way that returns the attributes and handle of the root dir.
Not doing so crashes nfsx because the inode table is not able to
find a inode for the root's parent. This inode was being referenced in
nfs3_lookup_parentdir_resume when filling a loc for the lookup fop.
For the record, such a lookup request is sent by vmkernel.
Signed-off-by: Shehjar Tikoo <shehjart@gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 942 (NFS crashes as a vmware ESX data store)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=942
In the readdir reply for the root of the export, replace the ino
and gen number for the . and .. entries with 1 and 0 respectively.
On clients which inspect this field, the client will error out due
to the change in inode number of the root directory when see for "."
.. also needs to be replaced because we do not have a concept of
the parent directory of root. The return of 1 and 0 is the same as
the behaviour of: stat /.. command.
Signed-off-by: Shehjar Tikoo <shehjart@gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 942 (NFS crashes as a vmware ESX data store)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=942
libglrpcsvc is a library and not a module according to Mac OS X.
Hence "-module" needs to be removed for linking to work.
Signed-off-by: Vikas Gorur <vikas@gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 361 (GlusterFS 3.0 should work on Mac OS/X)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=361
Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 945 (stat information of postparent stored in preparentstr instead of postparentstr in trace)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=945
Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 945 (stat information of postparent stored in preparentstr instead of postparentstr in trace)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=945
when client would send just the path as part of a transaction, path
reconstruction would create parent directory's path instead of actual
path at the time of resolution.
Signed-off-by: Anand V. Avati <avati@blackhole.gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 935 (Directories change mode from 0755 to 0644 automatically)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=935
- If there are non-contiguous offsets (offsets which do not start where
previous write ended), wait for completion of previous writes to server
before sending new ones.
- Send flush call to server only when all writes are completed.
- If a file is opened with O_APPEND, at any point of time a maximum only one
write call to server should be in transit. This is to avoid reordering of
writes in the presence of afr which can result in data corruption.
See bug #934 for more details.
Signed-off-by: Raghavendra G <raghavendra@gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 934 (md5sum mismatch when files are transferred using vsftpd)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=934
Introduces two new socket options:
1. transport.socket.keepalive: bool
Sets keepalive on a transport. On by default.
2. transport.socket.keepalive-interval: integer
Number of seconds between each keepalive message on the socket.
Default in linux is 2 hours. To change that value for a given
transport, use this option. Integer value is interval in seconds.
Signed-off-by: Shehjar Tikoo <shehjart@gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 754 (enable tcp keepalive)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=754
Introduces two new options:
1. nfs3.*.trusted-write: Forces UNSTABLE writes to return STABLE to NFS
clients to prevent the clients from sending a COMMIT. STABLE writes
are still handled in a sync manner and so are COMMITs if they're sent
at all.
2. nfs3.*.trusted-sync: Forces all WRITEs and COMMITs to return STABLE
return flags to NFS clients to avoid the overhead of STABLE writes, and
COMMITs that follow UNSTABLE writes. This includes the trusted-write
functionality. In addition to the trusted-write, it also writes
STABLE writes in an UNSTABLE manner.
Both violate the NFS protocol but allow better write perf in most
configurations. Use with caution.
Signed-off-by: Shehjar Tikoo <shehjart@gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 924 (Slow NFS synchronous writes)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=924
Injecting volumes at startup this way is pretty hacky but it will do for now.
Refactoring that part will come as separate commit.
Signed-off-by: Csaba Henk <csaba@gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 361 (GlusterFS 3.0 should work on Mac OS/X)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=361
Signed-off-by: lakshmipathi <lakshmipathi@gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 799 (dht prints 100% full when subvolume is not connected to remote host)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=799
dht - switch over to readdirp based algo if one subvolume is down
Signed-off-by: Anand V. Avati <avati@blackhole.gluster.com>
Signed-off-by: Anand V. Avati <avati@amp.gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 898 (small file performance enhancements)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=898
generation numbers should detect recreations and no need to rely on
layout changes anymore
Signed-off-by: Anand V. Avati <avati@blackhole.gluster.com>
Signed-off-by: Anand V. Avati <avati@amp.gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 898 (small file performance enhancements)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=898
Signed-off-by: Anand V. Avati <avati@blackhole.gluster.com>
Signed-off-by: Anand V. Avati <avati@amp.gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 898 (small file performance enhancements)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=898
The final unref on the inode during a file removal
should take place only if the file removal was successful.
Signed-off-by: Shehjar Tikoo <shehjart@gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 857 (Crash in afr_sh_entry_expunge_entry_cbk)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=857
glusterfs-profiler works in text mode by default now. This allows it to run on systems which don't have matplotlib installed.
Modes can be selected using:
-m graph - Graphical mode
-m text - Text mode (default)
Signed-off-by: Vikas Gorur <vikas@gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 268 (Add timing instrumentation code)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=268
Most of the time, a NFS request and the reply are separated by hundreds
of lines of other debugging output most of which is not always relevant
for NFS debugging. This script synthesizes the full glusterfs log into
the parts relevant for NFS while bringing together the NFS requests with
there replies on one line.
Min log-level required for this script to be useful is DEBUG. Only
works for NFS translator.
Signed-off-by: Shehjar Tikoo <shehjart@gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 913 (NFS-specific log analyzer tool)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=913
A previos commit brought in support for returning read replies
when subvolumes return reads in multiple iovecs. This did
not completely fix the problem since the bytes in iovecs
all together could be unaligned with the 4 byte
boundary as needed by XDR for the opaque data. This resulted
in read requests being either retransmitted or rejected with
an error message in syslog on the NFS client.
Signed-off-by: Shehjar Tikoo <shehjart@dev.gluster.com>
Signed-off-by: Shehjar Tikoo <shehjart@gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 902 (iozone hangs during random read throughput test)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=902
A previous change to this function introduced a regression
for exclusive creates. This patch fixes the regression by
bringing the return values in line with what the callers
expect in different situations of loc filling.
Signed-off-by: Shehjar Tikoo <shehjart@dev.gluster.com>
Signed-off-by: Shehjar Tikoo <shehjart@gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 857 (Crash in afr_sh_entry_expunge_entry_cbk)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=857
There is a possibility of io-cache or read-ahead returning
a read buffer that straddles two separate pages in ioc or ra,
through two struct iovecs. Current nfs3 read reply does not
return as many vectors as received from a subvolume leading to
a short read for the NFS client.
Signed-off-by: Shehjar Tikoo <shehjart@gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 902 (iozone hangs during random read throughput test)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=902
RPC reply path allows attaching a single iovec to a reply. This limits
the replies for certain program ops which need to return multiple vectors
for eg, the NFSv3 read reply.
Signed-off-by: Shehjar Tikoo <shehjart@gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 902 (iozone hangs during random read throughput test)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=902
Just because an inode_get for parent (ino, gen) returns a valid
inode_t does not meant that the inode_path(..in nfs_parent_inode_loc_fill..)
will return successfully. The failure of nfs_parent_inode_loc_fill was
not handled resulting in a NULL path being passed for the fop.
Signed-off-by: Shehjar Tikoo <shehjart@gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
BUG: 857 (Crash in afr_sh_entry_expunge_entry_cbk)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=857
