Time |
Nick |
Message |
00:22 |
|
wdennis joined #gluster |
00:33 |
|
roost joined #gluster |
00:51 |
|
wdennis joined #gluster |
00:52 |
wdennis |
hi all - going to use Gluster across two nodes (and possibly a third at some point) for an oVirt cluster's storage - if I utilize the native client, can the multiple nodes all share the same volumes? |
00:53 |
wdennis |
(I tyhink so, but just checking - the oVirt docs say to use NFS, but no need if I don't have to, since all the oVirt nodes will host bricks for the same Gluster volumes) |
01:05 |
Jmainguy |
wdennis: native gluster client? |
01:06 |
wdennis |
Jmainguy: yes |
01:06 |
Jmainguy |
wdennis: yeah that is no issue, you can even use nfs from gluster if you wanted to |
01:07 |
wdennis |
Jmainguy: just wanted to make sure, as the oVirt docs assume NFS mount |
01:07 |
Jmainguy |
wdennis: so you build your volume across multiple nodes, and then your clients mount the volumes, and many clients can mount one volume, via either gluster or nfs |
01:07 |
Jmainguy |
yeah just treat gluster like nfs for any nfs requirements |
01:07 |
wdennis |
Jmainguy: cool, thx |
01:07 |
Jmainguy |
you should be solid |
01:09 |
|
mikedep3- joined #gluster |
01:11 |
|
roost joined #gluster |
01:14 |
|
kdhananjay joined #gluster |
01:18 |
|
soumya joined #gluster |
01:46 |
|
julim joined #gluster |
01:46 |
|
msmith joined #gluster |
01:56 |
|
nangthang joined #gluster |
02:02 |
|
Anjana joined #gluster |
02:03 |
|
badone__ joined #gluster |
02:03 |
|
smohan joined #gluster |
02:06 |
|
roost joined #gluster |
02:10 |
|
harish joined #gluster |
02:36 |
|
smohan joined #gluster |
02:38 |
|
wkf joined #gluster |
02:42 |
|
shubhendu joined #gluster |
02:51 |
|
smohan_ joined #gluster |
02:55 |
misc |
JoeJulian: redirect what ? |
03:01 |
|
poornimag joined #gluster |
03:04 |
|
poornimag joined #gluster |
03:17 |
|
Guest33730 joined #gluster |
03:20 |
|
kumar joined #gluster |
03:20 |
|
kanagaraj joined #gluster |
03:37 |
|
jiku joined #gluster |
03:43 |
|
itisravi joined #gluster |
04:03 |
|
atinmu joined #gluster |
04:22 |
|
smohan joined #gluster |
04:23 |
|
pppp joined #gluster |
04:27 |
|
sakshi joined #gluster |
04:27 |
|
ppai joined #gluster |
04:30 |
|
anoopcs joined #gluster |
04:30 |
|
overclk joined #gluster |
04:30 |
|
Triveni joined #gluster |
04:36 |
|
vimal joined #gluster |
04:38 |
|
ashiq joined #gluster |
04:42 |
|
jiffin joined #gluster |
04:44 |
|
Manikandan joined #gluster |
04:48 |
|
iheartwhiskey joined #gluster |
04:53 |
|
sripathi joined #gluster |
04:54 |
|
ndarshan joined #gluster |
04:54 |
|
ashiq joined #gluster |
04:57 |
|
deepakcs joined #gluster |
04:57 |
|
rafi joined #gluster |
04:58 |
|
shubhendu joined #gluster |
05:07 |
|
wkf joined #gluster |
05:09 |
|
Bhaskarakiran joined #gluster |
05:11 |
|
ashiq joined #gluster |
05:14 |
|
schandra joined #gluster |
05:18 |
|
iheartwhiskey left #gluster |
05:18 |
|
iheartwhiskey joined #gluster |
05:19 |
|
lalatenduM joined #gluster |
05:19 |
|
raghu joined #gluster |
05:20 |
|
foster joined #gluster |
05:21 |
iheartwhiskey |
vhosts /Cloaks |
05:21 |
iheartwhiskey |
oops |
05:23 |
|
anil joined #gluster |
05:23 |
|
hagarth joined #gluster |
05:28 |
|
gem joined #gluster |
05:29 |
iheartwhiskey |
/msg NickServ SETPASS iheartwhiskey xjnkouoeecha re8THaqu |
05:29 |
iheartwhiskey |
/msg NickServ |
05:30 |
|
iheartwhiskey left #gluster |
05:30 |
|
SOLDIERz joined #gluster |
05:31 |
|
kshlm joined #gluster |
05:33 |
|
schandra joined #gluster |
05:33 |
|
maveric_amitc_ joined #gluster |
05:34 |
|
Philambdo joined #gluster |
05:37 |
|
plarsen joined #gluster |
05:43 |
|
rjoseph joined #gluster |
05:43 |
|
nbalacha joined #gluster |
05:46 |
|
aravindavk joined #gluster |
05:50 |
|
jcastill1 joined #gluster |
05:54 |
|
Guest33730 joined #gluster |
05:55 |
|
jcastillo joined #gluster |
05:57 |
|
foster joined #gluster |
05:59 |
|
DV joined #gluster |
05:59 |
|
atalur joined #gluster |
06:01 |
|
lalatenduM joined #gluster |
06:03 |
|
soumya joined #gluster |
06:05 |
|
DV joined #gluster |
06:07 |
|
gem joined #gluster |
06:13 |
|
liquidat joined #gluster |
06:19 |
|
jtux joined #gluster |
06:19 |
|
Pupeno joined #gluster |
06:19 |
|
Pupeno joined #gluster |
06:22 |
|
kotreshhr joined #gluster |
06:29 |
|
Anjana joined #gluster |
06:33 |
|
Guest33730 joined #gluster |
06:33 |
|
ghenry joined #gluster |
06:33 |
|
ghenry joined #gluster |
06:37 |
|
schandra joined #gluster |
06:42 |
|
aravindavk joined #gluster |
06:45 |
|
smohan_ joined #gluster |
06:47 |
|
hgowtham joined #gluster |
06:55 |
|
SOLDIERz joined #gluster |
07:07 |
|
hgowtham joined #gluster |
07:10 |
|
mbukatov joined #gluster |
07:11 |
|
Guest33730 joined #gluster |
07:15 |
|
lifeofguenter joined #gluster |
07:16 |
|
haomaiwa_ joined #gluster |
07:19 |
|
[Enrico] joined #gluster |
07:21 |
|
SOLDIERz joined #gluster |
07:22 |
glusterbot |
News from newglusterbugs: [Bug 1214169] glusterfsd crashed while rebalance and self-heal were in progress <https://bugzilla.redhat.com/show_bug.cgi?id=1214169> |
07:29 |
|
horacer joined #gluster |
07:32 |
|
DV joined #gluster |
07:42 |
|
smohan joined #gluster |
07:42 |
|
T0aD joined #gluster |
07:43 |
|
fsimonce joined #gluster |
07:46 |
|
gem joined #gluster |
07:46 |
|
deniszh joined #gluster |
07:47 |
|
deniszh left #gluster |
07:48 |
|
deniszh joined #gluster |
07:51 |
|
meghanam joined #gluster |
07:54 |
|
lifeofgu_ joined #gluster |
07:57 |
|
Slashman joined #gluster |
08:06 |
|
ctria joined #gluster |
08:20 |
|
ktosiek joined #gluster |
08:23 |
|
gem_ joined #gluster |
08:27 |
|
ghenry joined #gluster |
08:28 |
|
SOLDIERz joined #gluster |
08:30 |
|
Bhaskarakiran joined #gluster |
08:31 |
|
ndk joined #gluster |
08:32 |
|
Norky joined #gluster |
08:35 |
|
ppai joined #gluster |
08:35 |
|
sage_ joined #gluster |
08:36 |
|
karnan joined #gluster |
08:38 |
|
gem_ joined #gluster |
08:43 |
|
atalur joined #gluster |
08:44 |
|
mbukatov joined #gluster |
08:46 |
|
mbukatov joined #gluster |
08:47 |
|
al joined #gluster |
08:50 |
|
[Enrico] joined #gluster |
08:53 |
|
gem__ joined #gluster |
08:54 |
|
mbukatov joined #gluster |
08:56 |
|
mbukatov joined #gluster |
09:00 |
|
lifeofguenter joined #gluster |
09:16 |
|
kotreshhr1 joined #gluster |
09:16 |
|
anrao joined #gluster |
09:17 |
|
mbukatov joined #gluster |
09:17 |
|
Anjana joined #gluster |
09:17 |
|
mbukatov joined #gluster |
09:21 |
|
mbukatov joined #gluster |
09:22 |
|
Bardack joined #gluster |
09:25 |
|
ashiq joined #gluster |
09:26 |
|
poornimag joined #gluster |
09:27 |
|
hgowtham joined #gluster |
09:30 |
|
anrao joined #gluster |
09:33 |
|
ppai joined #gluster |
09:38 |
|
ira joined #gluster |
09:42 |
|
kdhananjay joined #gluster |
09:46 |
|
jcastill1 joined #gluster |
09:48 |
|
lifeofgu_ joined #gluster |
09:51 |
|
Manikandan joined #gluster |
09:51 |
|
Manikandan_ joined #gluster |
09:51 |
|
jcastillo joined #gluster |
09:52 |
glusterbot |
News from newglusterbugs: [Bug 1214220] Crashes in logging code <https://bugzilla.redhat.com/show_bug.cgi?id=1214220> |
09:52 |
glusterbot |
News from newglusterbugs: [Bug 1214222] Directories are missing on the mount point after attaching tier to distribute replicate volume. <https://bugzilla.redhat.com/show_bug.cgi?id=1214222> |
09:52 |
glusterbot |
News from newglusterbugs: [Bug 1214219] Data Tiering:Enabling quota command fails with "quota command failed : Commit failed on localhost" <https://bugzilla.redhat.com/show_bug.cgi?id=1214219> |
10:00 |
|
ira joined #gluster |
10:01 |
|
Pupeno joined #gluster |
10:11 |
|
Anjana joined #gluster |
10:22 |
|
spandit joined #gluster |
10:22 |
glusterbot |
News from newglusterbugs: [Bug 1214247] sharding - Implement remaining fops <https://bugzilla.redhat.com/show_bug.cgi?id=1214247> |
10:23 |
glusterbot |
News from newglusterbugs: [Bug 1214248] Persist file size and block count of sharded files in the form of xattrs <https://bugzilla.redhat.com/show_bug.cgi?id=1214248> |
10:23 |
glusterbot |
News from newglusterbugs: [Bug 1214249] I/O (fresh writes) failure of failure of hot tier <https://bugzilla.redhat.com/show_bug.cgi?id=1214249> |
10:23 |
|
gem__ joined #gluster |
10:34 |
|
raghug joined #gluster |
10:36 |
|
lifeofguenter joined #gluster |
10:36 |
|
harish_ joined #gluster |
10:38 |
|
anrao joined #gluster |
10:38 |
|
gem__ joined #gluster |
10:40 |
|
DV joined #gluster |
10:42 |
|
soumya_ joined #gluster |
10:53 |
glusterbot |
News from newglusterbugs: [Bug 1214273] detach-tier o/p needs refinement <https://bugzilla.redhat.com/show_bug.cgi?id=1214273> |
10:56 |
|
meghanam joined #gluster |
10:57 |
|
lifeofgu_ joined #gluster |
11:01 |
|
deepakcs joined #gluster |
11:08 |
|
gem__ joined #gluster |
11:09 |
|
meghanam_ joined #gluster |
11:11 |
|
firemanxbr joined #gluster |
11:11 |
|
SOLDIERz joined #gluster |
11:17 |
|
atalur joined #gluster |
11:20 |
|
ulimit joined #gluster |
11:23 |
glusterbot |
News from newglusterbugs: [Bug 1214289] I/O failure on attaching tier <https://bugzilla.redhat.com/show_bug.cgi?id=1214289> |
11:24 |
|
gem__ joined #gluster |
11:27 |
|
lyang0 joined #gluster |
11:42 |
|
DV__ joined #gluster |
11:45 |
|
jcastill1 joined #gluster |
11:47 |
|
LebedevRI joined #gluster |
11:48 |
|
jdarcy joined #gluster |
11:49 |
|
DV__ joined #gluster |
11:50 |
|
hgowtham joined #gluster |
11:50 |
|
jcastillo joined #gluster |
11:54 |
|
gem__ joined #gluster |
11:58 |
|
anoopcs joined #gluster |
11:59 |
|
anrao joined #gluster |
12:00 |
JustinClift |
*** Weekly Gluster Community Meeting is starting NOW on FreeNODE IRC #gluster-meeting *** |
12:01 |
|
hagarth1 joined #gluster |
12:04 |
|
side_control joined #gluster |
12:09 |
|
kotreshhr joined #gluster |
12:10 |
|
T0aD joined #gluster |
12:12 |
|
sahina joined #gluster |
12:13 |
|
soumya joined #gluster |
12:17 |
|
Gill_ joined #gluster |
12:18 |
|
rjoseph joined #gluster |
12:18 |
|
ppai joined #gluster |
12:23 |
|
kanagaraj joined #gluster |
12:28 |
|
rafi1 joined #gluster |
12:29 |
|
SOLDIERz joined #gluster |
12:29 |
|
chirino joined #gluster |
12:31 |
|
rafi joined #gluster |
12:31 |
|
B21956 joined #gluster |
12:39 |
|
gem__ joined #gluster |
12:41 |
|
shubhendu joined #gluster |
12:44 |
|
anil joined #gluster |
12:48 |
|
Anjana joined #gluster |
12:48 |
|
Philambdo joined #gluster |
12:51 |
|
kotreshhr left #gluster |
12:53 |
|
DV__ joined #gluster |
12:57 |
|
rjoseph joined #gluster |
12:58 |
|
Gill_ joined #gluster |
13:01 |
|
atalur joined #gluster |
13:05 |
|
rafi1 joined #gluster |
13:08 |
|
DV__ joined #gluster |
13:10 |
|
stickyboy joined #gluster |
13:12 |
|
plarsen joined #gluster |
13:18 |
|
dgandhi joined #gluster |
13:19 |
|
dgandhi joined #gluster |
13:21 |
|
dgandhi joined #gluster |
13:22 |
|
dgandhi joined #gluster |
13:23 |
|
dgandhi joined #gluster |
13:24 |
|
dgandhi joined #gluster |
13:25 |
|
dgandhi joined #gluster |
13:25 |
glusterbot |
News from resolvedglusterbugs: [Bug 1199303] GlusterFS 3.4.8 Tracker <https://bugzilla.redhat.com/show_bug.cgi?id=1199303> |
13:25 |
|
dgandhi joined #gluster |
13:26 |
|
dgandhi joined #gluster |
13:27 |
|
dgandhi joined #gluster |
13:27 |
|
bennyturns joined #gluster |
13:28 |
|
dgandhi joined #gluster |
13:29 |
|
dgandhi joined #gluster |
13:30 |
|
dgandhi joined #gluster |
13:30 |
|
georgeh-LT2 joined #gluster |
13:31 |
|
dgandhi joined #gluster |
13:31 |
|
rafi joined #gluster |
13:31 |
|
bene2 joined #gluster |
13:33 |
|
dgandhi joined #gluster |
13:34 |
|
dgandhi joined #gluster |
13:35 |
|
dgandhi joined #gluster |
13:36 |
|
DV__ joined #gluster |
13:37 |
|
dgandhi joined #gluster |
13:38 |
|
dgandhi joined #gluster |
13:40 |
|
dgandhi joined #gluster |
13:40 |
|
coredump joined #gluster |
13:41 |
|
dgandhi joined #gluster |
13:43 |
|
dgandhi joined #gluster |
13:44 |
|
dgandhi joined #gluster |
13:45 |
|
dgandhi joined #gluster |
13:45 |
|
jmarley joined #gluster |
13:46 |
|
dgandhi joined #gluster |
13:46 |
|
harish_ joined #gluster |
13:47 |
|
dgandhi joined #gluster |
13:47 |
|
msmith_ joined #gluster |
13:48 |
|
kanagaraj joined #gluster |
13:48 |
|
msmith_ joined #gluster |
13:48 |
|
dgandhi joined #gluster |
13:49 |
|
dgandhi joined #gluster |
13:50 |
|
dgandhi joined #gluster |
13:52 |
|
dgandhi joined #gluster |
13:53 |
glusterbot |
News from newglusterbugs: [Bug 1214339] Incorrect and unclear "vol info" o/p for tiered volume <https://bugzilla.redhat.com/show_bug.cgi?id=1214339> |
13:54 |
|
dgandhi joined #gluster |
13:55 |
|
dgandhi joined #gluster |
13:57 |
|
dgandhi joined #gluster |
13:58 |
|
dgandhi joined #gluster |
13:58 |
|
dgandhi joined #gluster |
13:59 |
|
dgandhi joined #gluster |
14:02 |
|
nangthang joined #gluster |
14:02 |
|
dgandhi joined #gluster |
14:02 |
|
bene3 joined #gluster |
14:03 |
|
neofob joined #gluster |
14:04 |
|
dgandhi joined #gluster |
14:06 |
|
dgandhi joined #gluster |
14:08 |
|
dgandhi joined #gluster |
14:09 |
|
dgandhi joined #gluster |
14:10 |
|
dgandhi joined #gluster |
14:12 |
|
dgandhi joined #gluster |
14:13 |
|
dgandhi joined #gluster |
14:13 |
|
wkf joined #gluster |
14:14 |
|
haomaiwa_ joined #gluster |
14:15 |
|
dgandhi joined #gluster |
14:17 |
|
dgandhi joined #gluster |
14:18 |
|
dgandhi joined #gluster |
14:18 |
|
dgandhi joined #gluster |
14:20 |
|
dusmant joined #gluster |
14:20 |
|
dgandhi joined #gluster |
14:21 |
|
SOLDIERz joined #gluster |
14:22 |
|
dgandhi joined #gluster |
14:22 |
|
wushudoin joined #gluster |
14:23 |
|
dgandhi joined #gluster |
14:25 |
|
dgandhi joined #gluster |
14:26 |
|
dgandhi joined #gluster |
14:27 |
|
SOLDIERz joined #gluster |
14:27 |
|
plarsen joined #gluster |
14:27 |
|
plarsen joined #gluster |
14:28 |
|
dgandhi joined #gluster |
14:29 |
|
dgandhi joined #gluster |
14:30 |
|
squizzi joined #gluster |
14:30 |
|
dgandhi joined #gluster |
14:32 |
|
dgandhi joined #gluster |
14:32 |
|
anrao joined #gluster |
14:34 |
|
dgandhi joined #gluster |
14:35 |
|
neofob joined #gluster |
14:35 |
|
dgandhi joined #gluster |
14:36 |
|
dgandhi joined #gluster |
14:36 |
|
jobewan joined #gluster |
14:38 |
|
dgandhi joined #gluster |
14:39 |
|
dgandhi joined #gluster |
14:40 |
|
dgandhi joined #gluster |
14:41 |
|
neofob joined #gluster |
14:42 |
|
dberry joined #gluster |
14:42 |
|
dberry joined #gluster |
14:42 |
|
dgandhi joined #gluster |
14:43 |
|
dgandhi joined #gluster |
14:44 |
|
dgandhi joined #gluster |
14:46 |
|
dgandhi joined #gluster |
14:47 |
|
coredump joined #gluster |
14:47 |
|
dgandhi joined #gluster |
14:48 |
|
dgandhi joined #gluster |
14:49 |
|
dgandhi joined #gluster |
14:51 |
partner |
umm it seems self-heal takes some much cpu it breaks the gluster and disconnects clients when bringing back the fixed hardware with its bricks, replica 2, gluster 3.4.5 on debian |
14:51 |
|
dgandhi joined #gluster |
14:52 |
partner |
any hints on how to tackle the issue? at the moment the self-heal daemon is disabled in order to resurrect the production |
14:52 |
|
dgandhi joined #gluster |
14:54 |
|
dgandhi joined #gluster |
14:54 |
|
Pupeno joined #gluster |
14:54 |
|
dgandhi joined #gluster |
14:55 |
partner |
obviously cannot manually self-heal in a controlled manner as the volume has the settings effective |
14:55 |
|
dgandhi joined #gluster |
14:56 |
|
dgandhi joined #gluster |
14:57 |
partner |
would/could it help if those different parts of heal would be enabled one at the time? metadata, data, entry,..? |
15:03 |
|
coredump joined #gluster |
15:03 |
|
roost joined #gluster |
15:06 |
|
virusuy_work joined #gluster |
15:07 |
|
Arminder- joined #gluster |
15:08 |
|
klaxa|work joined #gluster |
15:08 |
klaxa|work |
i think i asked this before, but we want to live-migrate with gfapi, so far unsuccessfully |
15:09 |
|
cholcombe joined #gluster |
15:09 |
klaxa|work |
oh wait |
15:09 |
klaxa|work |
no actually, this is the right channel, i think |
15:10 |
klaxa|work |
well the problem is that when we live-migrate a vm with libvirt from machine A to machine B, the vm keeps the gfapi connection to machine A |
15:11 |
klaxa|work |
we want the vm to connect to the gfapi on machine B though so we can shut down machine A |
15:11 |
|
wkf joined #gluster |
15:12 |
|
hagarth joined #gluster |
15:22 |
|
kshlm joined #gluster |
15:23 |
|
kdhananjay joined #gluster |
15:26 |
|
bene2 joined #gluster |
15:26 |
|
Prilly joined #gluster |
15:27 |
partner |
load is up to 30 |
15:29 |
|
coredump joined #gluster |
15:30 |
partner |
hmm maybe dropping the background-self-heal-count from default of 16.. |
15:39 |
bennyturns |
partner, what is going on? load is creeping up during a self heal? |
15:40 |
partner |
yeah, replica counterpart was offline and it seems when the missing bricks are brought online (with self-heal daemon enabled) the cpu goes sky-high and eventually things start to break |
15:41 |
|
T0aD joined #gluster |
15:41 |
partner |
just wondering the options here to decrease the load |
15:45 |
|
wdennis joined #gluster |
15:48 |
|
soumya joined #gluster |
15:54 |
glusterbot |
News from newglusterbugs: [Bug 1214387] Incorrect vol info post detach on disperse volume <https://bugzilla.redhat.com/show_bug.cgi?id=1214387> |
15:54 |
glusterbot |
News from newglusterbugs: [Bug 1214393] Cannot access volume (from client) post detach <https://bugzilla.redhat.com/show_bug.cgi?id=1214393> |
15:55 |
wdennis |
hi all -- added the requisite mount lines in /etc/fstab for my Gluster volumes (as per admin-guide > admin_settingup_clients.md) and when I boot the servers, the filesystems do not get mounted... |
15:56 |
wdennis |
If I log in and exec a 'mount -a' then they do successfully get mounted |
15:56 |
glusterbot |
News from resolvedglusterbugs: [Bug 1212822] Data Tiering: link files getting created on hot tier for all the files in cold tier <https://bugzilla.redhat.com/show_bug.cgi?id=1212822> |
15:57 |
|
deepakcs joined #gluster |
15:58 |
wdennis |
I assume the GlusterFS services are not running at mount time? What would I need to do to have the volumes mount successfully at boot time? |
16:01 |
wdennis |
I'm running CentOS 7.1 with GlusterFS 3.6.2 |
16:01 |
partner |
you're correct, its not exactly a new issue either, not sure what is the current kosher "fix" for that, some run mount -a from rc.local, some might tune something else.. |
16:04 |
wdennis |
partner: o.O Really? What to "tune"? (the 'run mount-a from rc.local' did occur to me, but HACK HACK HACK...) |
16:07 |
|
lifeofguenter joined #gluster |
16:07 |
partner |
dunno what is the proper fix for centos. the issue was described here when i bumped into it: https://joejulian.name/blog/glusterfs-volumes-not-mounting-in-debian-squeeze-at-boot-time/ |
16:08 |
wdennis |
Hopefully JoeJulian will see this and give me the official word... |
16:10 |
ira |
There's at least one bug in this area that I know of :/ |
16:12 |
|
lifeofgu_ joined #gluster |
16:12 |
wdennis |
partner: wow, from 2 yrs ago... Would have hoped that RHAT would have fixed on at least their distro by now... |
16:26 |
glusterbot |
News from resolvedglusterbugs: [Bug 1214393] Cannot access volume (from client) post detach <https://bugzilla.redhat.com/show_bug.cgi?id=1214393> |
16:26 |
wdennis |
ira: Seeing nothing in RHAT Bugzilla with search term 'GlusterFS mount boot'... Do you have a Bug #? |
16:29 |
|
haomaiw__ joined #gluster |
16:29 |
ira |
wdennis: I forget who has it... it deal with mount timing... |
16:30 |
ira |
Mainly that mount.glusterfs returns before the volume has fully mounted. |
16:30 |
JoeJulian |
I haven't seen that problem with Fedora. |
16:32 |
JoeJulian |
wdennis: check the audit log. Maybe it's an selinux bug. |
16:33 |
wdennis |
JoeJulian: selinux has been disabled on these hosts... But I will check the audit log nonetheless... |
16:34 |
|
nangthang joined #gluster |
16:39 |
|
wdennis joined #gluster |
16:39 |
wdennis |
JoeJulian: not sure exactly what I'm looking for in the audit.log, but I do see this seemingly abnormal message: |
16:39 |
wdennis |
type=ANOM_ABEND msg=audit(1429716384.944:60): auid=4294967295 uid=0 gid=0 ses=4294967295 pid=5106 comm="glusterfs" reason="memory violation" sig=11 |
16:44 |
|
maveric_amitc_ joined #gluster |
16:44 |
|
hagarth joined #gluster |
16:44 |
wdennis |
I also see this line in journalctl: |
16:44 |
wdennis |
Apr 22 11:26:21 ovirt-node-01 mount[1328]: /sbin/mount.glusterfs: line 13: /dev/stderr: No such device or address |
16:45 |
wdennis |
and immediately thereafter... |
16:45 |
wdennis |
Apr 22 11:26:21 ovirt-node-01 systemd[1]: glusterfs-data.mount mount process exited, code=exited status=1 |
16:45 |
JoeJulian |
interesting |
16:45 |
wdennis |
Apr 22 11:26:21 ovirt-node-01 systemd[1]: Unit glusterfs-data.mount entered failed state. |
16:48 |
JoeJulian |
wdennis: can you fpaste the client log please? |
16:49 |
wdennis |
JoeJulian: sorry, 'fpaste'? |
16:49 |
JoeJulian |
@paste |
16:49 |
glusterbot |
JoeJulian: For RPM based distros you can yum install fpaste, for debian, ubuntu, and arch it's pastebinit. Then you can easily pipe command output to [f] paste [binit] and it'll give you a URL. |
16:49 |
JoeJulian |
Or you can just open fpaste.org and copy/paste. |
16:50 |
JoeJulian |
I'm just lazy. |
16:50 |
|
meghanam joined #gluster |
16:52 |
wdennis |
JoeJulian: cool, learned something new today :) |
16:53 |
JoeJulian |
:) |
16:54 |
wdennis |
JoeJulian: http://ur1.ca/k85wc |
16:56 |
glusterbot |
News from resolvedglusterbugs: [Bug 1203776] Fix tiering bugs in DHT code. <https://bugzilla.redhat.com/show_bug.cgi?id=1203776> |
16:56 |
glusterbot |
News from resolvedglusterbugs: [Bug 1212400] Attach tier failing and messing up vol info <https://bugzilla.redhat.com/show_bug.cgi?id=1212400> |
16:58 |
|
jobewan joined #gluster |
16:58 |
JoeJulian |
wdennis: which file is that? That looks like glustershd.log. |
16:59 |
JoeJulian |
... but I could be wrong. I haven't actually looked at a tracelog in a while. |
16:59 |
JoeJulian |
Oh, wait... that's the cli log. |
16:59 |
JoeJulian |
cli_cmd_submit |
17:02 |
wdennis |
JoeJulian: yes, it was the cli.log |
17:04 |
wdennis |
JoeJulian: was that not the correct one? (nothing else with 'cli' in /var/log/glusterfs/ ) |
17:05 |
JoeJulian |
/var/log/glusterfs/$($mountpoint | tr '/' '-').log |
17:05 |
JoeJulian |
... more or less. |
17:09 |
wdennis |
JoeJulian: OK, try this one then... http://ur1.ca/k85z2 (/var/log/glusterfs/glusterfs-engine.log) |
17:14 |
|
Rapture joined #gluster |
17:22 |
JoeJulian |
wdennis: Oh, sure. That makes sense. You have enabled quorum and at the point you're trying to mount, it's not yet established. |
17:22 |
JoeJulian |
hmm |
17:25 |
|
hagarth joined #gluster |
17:26 |
glusterbot |
News from resolvedglusterbugs: [Bug 1202893] resource agent script for ganesha HA uses `service` to start/stop ganesha.nfsd <https://bugzilla.redhat.com/show_bug.cgi?id=1202893> |
17:26 |
glusterbot |
News from resolvedglusterbugs: [Bug 1206592] Data Tiering:Allow adding brick to hot tier too(or let user choose to add bricks to any tier of their wish) <https://bugzilla.redhat.com/show_bug.cgi?id=1206592> |
17:27 |
partner |
so umm, any thoughts on decreasing the load on self-heal situation? i would think the cpu is a bit too little for the box, at least haven't experienced similar stuff with the other hardware earlier |
17:27 |
partner |
ref "above couple of hours back" |
17:28 |
wdennis |
JoeJulian: situation was, both nodes were booting at the same time... |
17:29 |
wdennis |
If I just booted one or the other, the fstab mount would work? |
17:31 |
partner |
load is 20 now when only daemon is healing and client-side is disabled, everything barely stays up and alive now |
17:35 |
partner |
based on xattrop this should take 35 days during which time the files would be unaccessible.. |
17:36 |
wdennis |
OK, so if I boot any one system while the other is up, the Gluster filesystems do mount |
17:38 |
partner |
wdennis: replica volume? yes |
17:39 |
partner |
or doesn't even have to be that as long as any of the servers is reachable for the volume info |
17:42 |
wdennis |
partner: yes, repl=2 |
17:43 |
wdennis |
hold on - I did bounce the other one, and it *did not* mount the Gluster filesystems.. |
17:44 |
|
nangthang joined #gluster |
17:45 |
|
dusmant joined #gluster |
17:46 |
partner |
sounds like you are talking about the serverside, not the client side, nevermind me then |
17:53 |
wdennis |
wdennis: bounced same node again, and then it *did* mount correctly that time... seems a bit fragile... |
17:53 |
|
jackdpeterson joined #gluster |
17:58 |
wdennis |
ah well, #yolo... on with the show... |
17:59 |
|
shaunm joined #gluster |
18:03 |
|
klaxa joined #gluster |
18:09 |
|
_Bryan_ joined #gluster |
18:19 |
|
shaunm_ joined #gluster |
18:25 |
partner |
i'm still open for suggestions to tackle my issue (repeating myself as people come and go) |
18:27 |
JoeJulian |
wdennis: Sorry, I was dealing with a manufacturer's tech support. Another option for you would be to add a 3rd peer to allow your system to maintain server quorum. |
18:28 |
JoeJulian |
partner: cgroups? |
18:28 |
wdennis |
JoeJulian: yup, that's the plan, just haven't got there yet... |
18:41 |
partner |
JoeJulian: i have no idea how the gluster would react, possibly it would stay alive better or then not.. but thanks for the tip, having a look at that |
18:42 |
partner |
but need to figure out something, 250000 of the new files from the past 2-3 days are giving permission denied due to hw issues which lead to gluster issues |
18:43 |
|
lifeofguenter joined #gluster |
18:43 |
partner |
just feels a bit counterwise to limit when we are to my understanding way under needed capacity |
18:52 |
|
Arminder joined #gluster |
18:53 |
|
Arminder joined #gluster |
18:56 |
|
Arminder joined #gluster |
18:56 |
|
ricky-ticky joined #gluster |
18:56 |
|
Arminder joined #gluster |
19:00 |
|
Arminder joined #gluster |
19:02 |
|
theron joined #gluster |
19:02 |
|
marbu joined #gluster |
19:04 |
partner |
does self-heal daemon prioritize any heals or goes through its list from xattrop in an order? we've disabled the client-side triggered healings but yet anything returning empty files/permission denied are shortly functional files |
19:05 |
partner |
by prioritize i mean for example "hey that file just got accessed, i'll see into that soon" |
19:05 |
|
marbu joined #gluster |
19:10 |
|
mbukatov joined #gluster |
19:11 |
|
jobewan joined #gluster |
19:11 |
|
lkoranda joined #gluster |
19:15 |
|
neofob joined #gluster |
19:18 |
|
jcastill1 joined #gluster |
19:23 |
|
jcastillo joined #gluster |
19:25 |
|
Arminder joined #gluster |
19:25 |
|
Arminder joined #gluster |
19:37 |
|
rshade98 joined #gluster |
19:50 |
|
theron_ joined #gluster |
20:05 |
|
msmith joined #gluster |
20:31 |
|
lexi2 joined #gluster |
20:46 |
|
wkf joined #gluster |
20:48 |
|
theron joined #gluster |
20:53 |
|
ZakWolfinger joined #gluster |
20:56 |
|
sage__ joined #gluster |
20:57 |
ZakWolfinger |
Gluster 3.5.3 servers and client, volume is a replica 2. When tar'ing a directory tree from the client I'm getting random "File removed before we read it" and if I try to copy the tree I get random "No such file or directory". Looking on the servers, the files exist. Volume heal gv0 info shows nothing needing healed. Any thoughts? |
21:00 |
|
sage_ joined #gluster |
21:09 |
|
msmith joined #gluster |
21:22 |
|
badone_ joined #gluster |
21:41 |
|
theron joined #gluster |
21:46 |
|
neofob joined #gluster |
21:53 |
partner |
i guess it will run over night now, we will attempt Something(tm) tomorrow, loads of files not served still |
21:53 |
partner |
daemon takes care of roughly 300 files per hour :( |
21:55 |
partner |
and despite the "client won't trigger any heal" an access to a file will somehow get it shortly fixed, not on the fly but next attempt will succeed, i recall seeing a bug related |
21:56 |
partner |
its ok, its not aggressive so its not bringing down the storage now, load seems to be steadily at 25 but clients get served |
21:57 |
JoeJulian |
partner: You should write something up about this. I suspect there's a lot of people that would like to do what you're doing. |
21:58 |
partner |
i'm just guessing :) |
22:01 |
partner |
i would love to share everything but it just seems i never make it to the blog post level, "this" is my blog where i share the stuff |
22:06 |
partner |
JoeJulian: but i do appreciate your persistance kicking me forward, unfortunately i seem to always disappoint you and the community on the topic :/ |
22:08 |
partner |
and as said earlier, i'm no longer maintaining this infra, just lending a helping hand to my former team and colleagues, this is just all new issue to me but perhaps the more the merrier :o |
22:14 |
partner |
we have half of dozen of ideas on how to proceed but even if we can manage to fix 1 file per 1 second it will still take some 70 hours to get done.. best would be if we could allow clients to trigger the heal but as for now its impossible due to load issues |
22:15 |
partner |
we just might try adding physical cpu (one socket free, donors available), perhaps try cutting down clients while still keeping production running, playing with various options available,... |
22:16 |
partner |
any and every attempt will anyways take an hour or so to level up and realisticly see the effect for the better or worse |
22:39 |
|
PaulCuzner joined #gluster |
22:53 |
|
dgandhi joined #gluster |
23:01 |
|
Rapture joined #gluster |
23:03 |
|
deniszh joined #gluster |
23:12 |
|
jvandewege joined #gluster |
23:15 |
|
edwardm61 joined #gluster |
23:27 |
|
msmith joined #gluster |
23:46 |
|
gildub joined #gluster |