Perl 6 - the future is here, just unevenly distributed

IRC log for #gluster-dev, 2017-08-18

| Channels | #gluster-dev index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:04 shyam joined #gluster-dev
01:03 sahina joined #gluster-dev
01:14 gyadav joined #gluster-dev
01:37 sac joined #gluster-dev
01:51 ilbot3 joined #gluster-dev
01:51 Topic for #gluster-dev is now Gluster Development Channel - https://www.gluster.org | For general chat go to #gluster | Patches - https://review.gluster.org/ | Channel Logs - https://botbot.me/freenode/gluster-dev/ & http://irclog.perlgeek.de/gluster-dev/
02:27 Shu6h3ndu joined #gluster-dev
02:31 pkalever joined #gluster-dev
02:47 Shu6h3ndu_ joined #gluster-dev
03:25 aravindavk joined #gluster-dev
03:31 gyadav joined #gluster-dev
03:33 ppai joined #gluster-dev
03:38 riyas joined #gluster-dev
03:39 aravindavk joined #gluster-dev
03:48 nbalacha joined #gluster-dev
03:51 itisravi joined #gluster-dev
04:00 hgowtham joined #gluster-dev
04:26 jiffin joined #gluster-dev
04:35 atinmu joined #gluster-dev
04:43 amarts joined #gluster-dev
04:56 Humble joined #gluster-dev
05:08 ndarshan joined #gluster-dev
05:08 sahina joined #gluster-dev
05:10 gyadav_ joined #gluster-dev
05:12 poornima_ joined #gluster-dev
05:14 rafi joined #gluster-dev
05:18 susant joined #gluster-dev
05:22 skumar joined #gluster-dev
05:23 gobindadas joined #gluster-dev
05:23 gyadav__ joined #gluster-dev
05:31 skumar_ joined #gluster-dev
05:32 sona joined #gluster-dev
05:32 skumar_ joined #gluster-dev
05:38 hgowtham joined #gluster-dev
05:40 hgowtham joined #gluster-dev
05:46 skumar__ joined #gluster-dev
05:47 kotreshhr joined #gluster-dev
05:47 msvbhat joined #gluster-dev
05:49 atinmu joined #gluster-dev
05:50 susant left #gluster-dev
05:58 Saravanakmr joined #gluster-dev
05:59 kdhananjay joined #gluster-dev
06:02 apandey joined #gluster-dev
06:05 atinmu joined #gluster-dev
06:06 kotreshhr left #gluster-dev
06:21 rastar joined #gluster-dev
06:29 kotreshhr joined #gluster-dev
06:31 atinmu joined #gluster-dev
06:34 pranithk1 joined #gluster-dev
06:35 rafi1 joined #gluster-dev
06:36 Venkata joined #gluster-dev
06:36 skoduri joined #gluster-dev
06:45 nbalacha jiffin, ping
06:52 misc nigelb: the gerrit stage server was full, I extended the / partition by 1G
06:53 gyadav_ joined #gluster-dev
07:17 gyadav__ joined #gluster-dev
07:22 kotreshhr joined #gluster-dev
07:28 ankitr joined #gluster-dev
07:32 ankitr joined #gluster-dev
07:41 kotreshhr left #gluster-dev
07:54 nbalacha anoopcs, ping
08:11 nbalacha anoopcs, jiffin , how does rebalance affect the trash translator?
08:15 ankitr joined #gluster-dev
08:39 anoopcs nbalacha, It basically checks for the file which is getting moved to the newly added brick.
08:40 anoopcs i.e, file is expected to be in trash directory of the source.
08:40 nigelb misc: thanks
08:40 misc nigelb: I am pondering to put /review on a separate partition
08:40 misc I am not sure what prevent it
08:41 misc (likely some missing feature I am waiting to merge in ansible since 5 months as usual)
08:41 Saravanakmr joined #gluster-dev
08:42 anoopcs nbalacha, I think the patch wrongly replaces the name of file which is being checked.
08:43 nbalacha anoopcs, the test seems to assume that a particular file is going to be migrated
08:43 nbalacha the rebalance changes seems to have switched the layout so the other one is getting migrated now
08:43 anoopcs nbalacha, Yes.
08:44 nbalacha anoopcs, I am goin to modify the test to find out which file was migrated and proceed accordingly
08:44 anoopcs nbalacha, Is that a change which is being introduced with this patch?
08:44 anoopcs I mean the layout change?
08:46 nbalacha anoopcs, the layout change seems to be a sideeffect of the changes
08:46 nbalacha it has basically flipped it
08:47 anoopcs nbalacha, Did you try running the test with the old file name? i.e, rebal2
08:47 nbalacha anoopcs, yes. it fails consistently as it is now rebal1 which is migrated to brick3
08:48 anoopcs nbalacha, Then rebal1 should work fine...
08:48 nbalacha anoopcs, it will but I dont want to hardcode it
08:49 nbalacha it is better to find out which file migrated and use that name
08:49 anoopcs Oh..Ok
08:50 nbalacha anoopcs, thank you. I will refresh the patch. Can you take a look after say 10 mins
08:51 amarts nbalacha++ on fixing that trash.t
08:51 glusterbot amarts: nbalacha's karma is now 10
08:52 amarts i was always shocked to see how we are always sure the specific file will be migrated
08:55 ashiq joined #gluster-dev
09:04 msvbhat joined #gluster-dev
09:11 sahina joined #gluster-dev
09:16 nigelb misc: crap, I keep forgetting we haven't closed that out.
09:18 misc nigelb: isn't why we want to have a high bandwidht hacking session in $undecided_location ?
09:18 misc (so we can be faster at forgetting it :p )
09:19 nigelb Indeed.
09:19 nigelb Or rather so we can break it and fix it immediately.
09:25 amarts poornima_, ping
09:26 amarts poornima_, when are you planning to look at gfproxy (TSP only) patch to go out? (or patch to pass regression)?
09:28 anoopcs nbalacha, Sorry.. was afk for a while. Will look into the updated patch set.
09:28 nbalacha anoopcs, np. Thanks
09:32 rtjure joined #gluster-dev
09:59 sona joined #gluster-dev
10:15 hgowtham joined #gluster-dev
10:28 itisravi joined #gluster-dev
10:47 amarts joined #gluster-dev
10:55 aravindavk joined #gluster-dev
11:14 rafi2 joined #gluster-dev
11:15 rafi2 ping bfoster
11:16 bfoster rafi2: pong
11:28 obnox joined #gluster-dev
11:39 rafi bfoster: we had problem with time attributes with gluster ,
11:40 rafi bfoster: different replica/ distribution set can have different time attributes
11:40 rafi bfoster: so thought of discussing with you
11:40 rafi bfoster: do you have some time ?
11:40 gyadav_ joined #gluster-dev
11:42 rafi bfoster: this is the issue i'm talking about https://github.com/gluster/glusterfs/issues/208
11:45 bfoster rafi: hm.. sounds like that is more of a problem with picking/using ctimes from different servers/bricks..?
11:48 ankitr joined #gluster-dev
11:48 rafi bfoster: true, now let us you served the ctime from one replica set,
11:49 riyas joined #gluster-dev
11:49 rafi bfoster: if that particular replica is down , you may need to serve from the available replica's
11:49 rafi bfoster: it could be different , and application will get inconsistent data
11:50 hgowtham joined #gluster-dev
11:52 bfoster yeah, not clear to me there are any great options outside of what pranith suggests in that mail
11:53 rafi bfoster: coool, So I was trying out that,
11:54 rafi bfoster: is there any good document about how the files systems like xfs /zfs update the time
11:55 rafi bfoster: main doubts regarding when an fsync happens, w
11:56 bfoster heh, not really. I'm not really sure myself without digging into the code :P
11:57 bfoster it looks like it actually may be a generic/vfs thing, which makes sense
11:57 bfoster (as far triggering when time updates occur)
11:57 bfoster as far as*
12:00 rafi bfoster: okey , I can look into the code,
12:00 bfoster see file_update_time() in fs/inode.c
12:00 rafi bfoster: it has more complexity when it comes to different mount options related to time
12:02 rafi bfoster: i think that is a good point to start
12:03 rafi bfoster: when an fsync happens can we know what are the things need to be synced , it may contain inode update/data update etc right ?
12:03 rafi bfoster: so when a sync happens it may or may not update the times, it depends on what need to be synced
12:03 rafi ?
12:04 bfoster I think fsync is more of a checkpoint operation than one that updates inode times, but I could be mistaken
12:04 bfoster e.g., other operations (such as writes) update the time on the in-core inode
12:04 bfoster fsync just forces the in-core inode to disk
12:05 uebera|| Hi. I've deinstalled the packages glusterfs-{common,client,server}-3.10.5-ubuntu1~xenial1 on Ubuntu 16.04, moved /var/lib/glusterd elsewhere and then reinstalled the three packages. After probing peer2 from peer1, I see that both have "operating-version=31004" in /var/lib/glusterd/glusterd.info. Isn't that supposed to be "31005"?
12:14 rafi joined #gluster-dev
12:14 rafi bfoster: great, that explanation clears my doubt
12:17 rafi joined #gluster-dev
12:23 aravindavk joined #gluster-dev
12:31 jstrunk joined #gluster-dev
12:36 jstrunk joined #gluster-dev
12:37 kotreshhr joined #gluster-dev
13:45 gyadav joined #gluster-dev
13:52 shyam joined #gluster-dev
13:58 gyadav_ joined #gluster-dev
14:03 susant joined #gluster-dev
14:05 gyadav joined #gluster-dev
14:12 baojg joined #gluster-dev
14:29 rastar joined #gluster-dev
14:45 skumar joined #gluster-dev
14:53 rraja joined #gluster-dev
14:55 ndevos uebera||: the op-version only gets updated in the sources when there is a need for it, 3.10.5 wont have any changes that requires a higher op-version
15:07 wushudoin joined #gluster-dev
15:08 Shu6h3ndu_ joined #gluster-dev
15:09 wushudoin joined #gluster-dev
15:20 vbellur joined #gluster-dev
15:31 nbalacha joined #gluster-dev
15:59 uebera|| ndevos: I see. Thanks!
16:09 gyadav_ joined #gluster-dev
16:19 vbellur joined #gluster-dev
16:23 aravindavk joined #gluster-dev
16:53 riyas joined #gluster-dev
16:58 atinmu joined #gluster-dev
17:22 sona joined #gluster-dev
17:35 Humble joined #gluster-dev
17:46 Humble joined #gluster-dev
18:05 shyam joined #gluster-dev
18:10 vbellur joined #gluster-dev
19:13 rastar joined #gluster-dev
19:13 rafi1 joined #gluster-dev
19:46 rafi joined #gluster-dev
19:51 vbellur joined #gluster-dev

| Channels | #gluster-dev index | Today | | Search | Google Search | Plain-Text | summary