Re: [Gluster-devel] Meaning of various log files

[jayakrishnan mm]

Thanks Kaushal. I saw  this  document. It is helpful

Best regards
JK

On Tue, Jun 14, 2016 at 12:04 PM, Kaushal M  wrote:

> On Fri, Jun 10, 2016 at 1:19 PM, jayakrishnan mm
>  wrote:
> > Hi ,
> >
> > I see  some  overlapping information between
> > /var/log/glusterfs/glustershd.log  and
> >
> > /var/log/glusterfs/mnt-gluster-.log. What is the difference
> > between these logs?
> >
> > volume name is ec-vol
> >
> >
>
> You can refer to the docs for this,
> https://gluster.readthedocs.io/en/latest/Administrator%20Guide/Logging/
>
> >
> >
> > Best regards
> > JK
> >
> > ___
> > Gluster-devel mailing list
> > Gluster-devel@gluster.org
> > http://www.gluster.org/mailman/listinfo/gluster-devel
>
___
Gluster-devel mailing list
Gluster-devel@gluster.org
http://www.gluster.org/mailman/listinfo/gluster-devel

[Gluster-devel] ./tests/basic/afr/self-heald.t regression failure

[Poornima Gurusiddaiah]

Hi, 

The test ./tests/basic/afr/self-heald.t creates a core and fails test 38, 53, 
68. 
Link for the regression 
https://build.gluster.org/job/rackspace-netbsd7-regression-triggered/17536/console
 

Can you please take a look at it? 

Regards, 
Poornima 
___
Gluster-devel mailing list
Gluster-devel@gluster.org
http://www.gluster.org/mailman/listinfo/gluster-devel

Re: [Gluster-devel] Meaning of various log files

[Kaushal M]

On Fri, Jun 10, 2016 at 1:19 PM, jayakrishnan mm
 wrote:
> Hi ,
>
> I see  some  overlapping information between
> /var/log/glusterfs/glustershd.log  and
>
> /var/log/glusterfs/mnt-gluster-.log. What is the difference
> between these logs?
>
> volume name is ec-vol
>
>

You can refer to the docs for this,
https://gluster.readthedocs.io/en/latest/Administrator%20Guide/Logging/

>
>
> Best regards
> JK
>
> ___
> Gluster-devel mailing list
> Gluster-devel@gluster.org
> http://www.gluster.org/mailman/listinfo/gluster-devel
___
Gluster-devel mailing list
Gluster-devel@gluster.org
http://www.gluster.org/mailman/listinfo/gluster-devel

[Gluster-devel] Brick multiplexing approaches

[Jeff Darcy]

"Brick multiplexing" is a new feature, tentatively part of 4.0, that
allows multiple bricks to be served from a single glusterfsd process.
This promises to give us many benefits over the current "process per
brick" approach.

 * Lower total memory use, by having only one copy of various global
   structures instead of one per brick/process.

 * Less CPU contention.  Every glusterfsd process involves several
   threads.  If there are more total threads than physical cores, or if
   those cores are also needed for other work on the same system, we'll
   thrash pretty badly.  As with memory use, managing each thread type 
   as a single pool (instead of one pool per brick/process) will help.

 * Fewer ports.  In the extreme case, we need only have one process and
   one port per node.  This avoids port exhaustion for high node/brick
   counts, and can also be more firewall-friendly.

 * Better coordination between bricks e.g. to implement QoS policies.

In short, our current infrastructure just isn't going to let us keep up
with various trends - higher node counts, containers, hyperconvergence,
even erasure coding.  The question is: how do we do it?  There are two
basic models.

 * In the "multiple graph" model, we have multiple separate graphs
   (volfiles) in a single process.  This allows them to share logging
   data and threads, polling threads, and many other resources -
   everything anchored at a glusterfs_ctx_t.  It does *not* allow them 
   to share ports, or anything else anchored by a protocol/server
   translator instance.

 * In the "single graph" model, we have multiple graphs joined together
   at the single protocol/server translator.  This allows more things to
   be shared, including ports, but does introduce some new problems.
   For one thing, it doesn't work if the bricks have different transport
   characteristics (e.g. TLS vs. non-TLS).  For another, it raises the
   possibility of a configuration change for one brick causing a graph
   switch that affects all bricks in the process (even if they belong to
   separate volumes).

I'd prefer to implement the single-graph model, because both the port 
conservation/exhaustion and QoS-coordination issues are important.
However, that means solving some of the additional problems.  In
particular...

 * Most of the incompatible-transport issues can be solved by moving
   various things from the server translator's "private" structure
   (server_conf_t) into per-connection or per-tenant structures.  I've 
   already done something similar with the inode table for subvolume
   mounts (#13659 in Gerrit) and it's a pain but it's feasible.  We
   might also (eventually) need consider implementing parts of the
   multi-graph model as well to host bricks in the same process even
   when their transports are incompatible.

 * For the graph-switch problem, we'll need to introduce some idea of
   sub-graphs or related graphs, so that we can compare and switch only
   the part relevant to a single brick.  I'd actually like to avoid this
   entirely until we get to GlusterD 2.0, but I'm not sure if we'll be
   able to get away with that.

Suggestions, warnings, or other thoughts are welcome.
___
Gluster-devel mailing list
Gluster-devel@gluster.org
http://www.gluster.org/mailman/listinfo/gluster-devel

[Gluster-devel] Replacing a host using gdeploy

[Sachidananda URS]

Hi,

With the replace node operation being very tedious and elaborate process,
it is
easy to commit mistakes. And it is much tedious process to roll back in
case of
mistakes.

This is an effort to automate the entire process, the script
replace_node.sh is
a wrapper script around gdeploy configuration file. It takes three
parameters:
`Existing node name', `new node name', and `volume name'. The script
generates
a configuration file, and calls gdeploy with the configuration file as an
argument. This handles clusters with multiple hosts and just two hosts.

USAGE:

$ ./replace_node.sh NODE-1 NODE-2 VOLUMENAME

***
Node NODE-2 will be replaced with configuration from NODE-1
It is very important to give the node names right. Review before continuing.
***

Continue(y/N):


-

When continued NODE-2 will be added into the cluster and self-heal is
triggered.

Prerequisites:
===
1. Password less ssh from node-1 to node-2
2. Passwordless ssh from the laptop to node-1, node-2
3. Place the replace_prep.sh under /usr/share/ansible/gdeploy/scripts/

Configuration generation and additional script generation and invocation is
managed by gdeploy.

The step-3 in Prerequisite will be handled in the rpm once the code is
merged
upstream and rpm generated. I'll post on -users list once I build rpms.

I've done some initial tests and things work fine. Any suggestions and
contributions are welcome.

-sac


replace_node.sh
Description: Bourne shell script


replace_node.conf.template
Description: Binary data


replace_prep.sh
Description: Bourne shell script
___
Gluster-devel mailing list
Gluster-devel@gluster.org
http://www.gluster.org/mailman/listinfo/gluster-devel

[Gluster-devel] Bugs with incorrect status

[Niels de Vos]

Date: Mon, 13 Jun 2016 09:06:53 +

1265893 (mainline) ON_QA: Perf: Getting bad performance while doing ls
  [master] If2e1a2 dht: improving readdir performance (NEW)
  [master] I716d48 storage/posix: Reduce number of getxattrs for internal 
xattrs (MERGED)
  [master] I8d1b2a storage/posix: Prevent extra handle-path (MERGED)
  ** pkara...@redhat.com: Bug 1265893 should be in POST, change If2e1a2 under 
review **

1099683 (3.5.0) POST: Silent error from call to realpath in 
features/changelog/lib/src/gf-history-changelog.c
  ** vshan...@redhat.com: No change posted, but bug 1099683 is in POST **

1332074 (3.7.10) MODIFIED: Marker: Lot of dict_get errors in brick log!!
  [release-3.7] I8054ff features/marker: Fix dict_get errors when key is NULL 
(MERGED)
  ** khire...@redhat.com: Bug 1332074 should be ON_QA, use v3.7.12rc1 for 
verification of the fix **

1294675 (3.7.6) MODIFIED: Healing queue rarely empty
  [release-3.7] If7eee1 cluster/afr: Fix spurious entries in heal info (MERGED)
  ** pkara...@redhat.com: Bug 1294675 should be ON_QA, use v3.7.12rc1 for 
verification of the fix **

1209329 (mainline) ON_QA: glusterd services are not handled properly when re 
configuring services
  [master] Ibd1cfd glusterd: Modify reconfigure function for glusterd services 
(NEW)
  [master] I6db10c glusterd: bitd daemon should stop after doing volume reset 
 force (ABANDONED)
  [master] I852c46 glusterd: Stop/restart/notify to daemons(svcs) during 
reset/set on a volume (MERGED)
  [master] I9ebe78 glusterd: Stop or restart svc services during reset on a 
volume. (ABANDONED)
  [master] I1c96c8 glusterd: Stop/restart svc services during reset on a 
volume. (ABANDONED)
  ** anekk...@redhat.com: Bug 1209329 should be in POST, change Ibd1cfd under 
review **

1316808 (3.7.8) MODIFIED: Data Tiering:tier volume status shows as in-progress 
on all nodes of a cluster even if the node is not part of volume
  [release-3.7] Ie4345b Tier/glusterd: Resetting the tier status value to not 
started (MERGED)
  [release-3.7] I15399d Tier: displaying status only one the nodes running 
tierd (MERGED)
  ** hgowt...@redhat.com: Bug 1316808 should be ON_QA, use v3.7.12rc1 for 
verification of the fix **

1258144 (3.7.5) ON_QA: Data Tiering: Tier deamon crashed when detach tier start 
was issued while IOs were happening
  ** dlamb...@redhat.com: No change posted, but bug 1258144 is in ON_QA **

1063506 (3.7.0) ASSIGNED: No xml output on gluster volume heal info command 
with --xml
  [release-3.7] I64c4ae heal/xml : xml implementation of heal info and 
splitbrain info (MERGED)
  ** pkara...@redhat.com: Bug 1063506 should be ON_QA, use v3.7.12rc1 for 
verification of the fix **

1339149 (mainline) POST: Error and warning messages related to 
xlator/features/snapview-client.so adding up to the client log on performing IO 
operations
  [master] Ib6ddaa snapshot/uss: client-side log is filled with 
"invalid-argument" error (MERGED)
  [master] I5c3f17 snapshot/uss: client log is filled with "invalid argument: 
inode" error (MERGED)
  [master] I8f7cc0 snapshot/uss: log is filled with gfid is NULL error (MERGED)
  ** rjos...@redhat.com: Bug 1339149 should be MODIFIED, change I8f7cc0 has 
been merged **

1209484 (mainline) ON_QA: Unable to stop/start a volume
  ** rkavu...@redhat.com: No change posted, but bug 1209484 is in ON_QA **

1326212 (3.7.10) ASSIGNED: gluster volume heal info shows conservative merge 
entries as in split-brain
  [release-3.7] I5d9a6d cluster/afr: Fix witness counting code in src/sink 
detection (MERGED)
  ** pkara...@redhat.com: Bug 1326212 should be ON_QA, use v3.7.12rc1 for 
verification of the fix **

1336199 (3.7.11) MODIFIED: failover is not working with latest builds.
  [release-3.7] Iabd06e common-ha: floating IP (VIP) doesn't fail over when 
ganesha.nfsd dies (MERGED)
  ** kkeit...@redhat.com: Bug 1336199 should be ON_QA, use v3.7.12rc1 for 
verification of the fix **

1333241 (3.7.11) MODIFIED: Fix excessive logging due to NULL dict in dht
  [release-3.7] I443322 cluster/dht: Perform NULL check on xdata before 
dict_get() (MERGED)
  ** b...@gluster.org: Bug 1333241 should be ON_QA, use v3.7.12rc1 for 
verification of the fix **

1296689 (mainline) ASSIGNED: Use cluster.watermark-hi to throttle creates on 
the hot tier.
  [master] I78b8f7 cluster/tier: WIP have watermark throttle creates on hot 
tier (NEW)
  ** dlamb...@redhat.com: Bug 1296689 should be in POST, change I78b8f7 under 
review **

1334441 (3.7.11) MODIFIED: SAMBA-VSS : Permission denied issue while restoring 
the directory from windows client 1 when files are deleted from windows client 2
  [release-3.7] Ib192ad gfapi: clear loc.gfid when retrying after ESTALE 
(MERGED)
  ** rta...@redhat.com: Bug 1334441 should be ON_QA, use v3.7.12rc1 for 
verification of the fix **

1344885 (mainline) NEW: inode leak in brick process
  [master] I565e81 libglusterfs/client_t: Dump the 0th client too (NEW)
  ** b...@gluster.org: Bug 1344885 should be in POST, change