Perl 6 - the future is here, just unevenly distributed

IRC log for #gluster, 2015-12-16

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

All times shown according to UTC.

Time Nick Message
00:09 DV joined #gluster
00:26 gildub joined #gluster
00:30 zhangjn joined #gluster
00:32 Intensity I'm also wondering if GlusterFS can somehow export or present (Mac OS X) extended attributes and resource forks, either by installing an OS X GlusterFS client or by exporting a (NFS/AFP/other?) filesystem to OS X that has GlusterFS as its underlying filesystem.
00:33 DV joined #gluster
00:44 MACscr joined #gluster
00:45 MACscr joined #gluster
00:46 cjellick_ joined #gluster
00:48 plarsen joined #gluster
00:53 m0zes joined #gluster
00:58 EinstCrazy joined #gluster
01:26 Lee1092 joined #gluster
01:47 amye joined #gluster
01:57 _Bryan_ joined #gluster
01:59 jrm16020 joined #gluster
02:02 rjoseph joined #gluster
02:16 zhangjn joined #gluster
02:19 zhangjn joined #gluster
02:19 nangthang joined #gluster
02:30 zhangjn joined #gluster
02:52 zhangjn joined #gluster
02:58 zhangjn joined #gluster
02:58 shubhendu joined #gluster
03:04 zhangjn joined #gluster
03:07 zhangjn joined #gluster
03:13 cjellick joined #gluster
03:19 zhangjn joined #gluster
03:20 nishanth joined #gluster
03:28 atinm joined #gluster
03:34 cjellick joined #gluster
03:43 ashiq joined #gluster
03:43 kanagaraj joined #gluster
03:43 nbalacha joined #gluster
03:43 Peppaq joined #gluster
03:51 harish_ joined #gluster
03:54 suliba joined #gluster
03:57 hagarth joined #gluster
04:01 RameshN joined #gluster
04:05 kshlm joined #gluster
04:05 ahino joined #gluster
04:05 MACscr joined #gluster
04:06 MACscr joined #gluster
04:06 itisravi joined #gluster
04:07 amye joined #gluster
04:09 suliba joined #gluster
04:13 hgichon joined #gluster
04:15 rjoseph joined #gluster
04:21 ramteid joined #gluster
04:27 Apeksha joined #gluster
04:30 zhangjn joined #gluster
04:36 Manikandan joined #gluster
04:37 kotreshhr joined #gluster
04:44 RameshN joined #gluster
04:45 nehar joined #gluster
04:46 calavera joined #gluster
04:58 ndarshan joined #gluster
05:03 ppai joined #gluster
05:22 Bhaskarakiran joined #gluster
05:24 itisravi joined #gluster
05:26 pppp joined #gluster
05:27 kdhananjay joined #gluster
05:29 vmallika joined #gluster
05:31 skoduri joined #gluster
05:32 zhangjn joined #gluster
05:35 ahino joined #gluster
05:38 Apeksha joined #gluster
05:48 poornimag joined #gluster
05:53 jiffin joined #gluster
05:54 suliba joined #gluster
05:56 arcolife joined #gluster
05:59 anil joined #gluster
06:00 rafi joined #gluster
06:01 overclk joined #gluster
06:04 ashiq joined #gluster
06:07 ahino joined #gluster
06:11 hgowtham joined #gluster
06:14 deepakcs joined #gluster
06:21 shubhendu joined #gluster
06:28 wtracz2 joined #gluster
06:28 wtracz2 left #gluster
06:36 SOLDIERz joined #gluster
06:42 DV joined #gluster
06:44 itisravi joined #gluster
07:08 nangthang joined #gluster
07:16 atinm joined #gluster
07:18 cjellick joined #gluster
07:20 mhulsman joined #gluster
07:24 overclk joined #gluster
07:31 jtux joined #gluster
07:33 klaxa joined #gluster
07:42 MACscr is it me or does gluster have tons of issues consistently mounting a client volume a boot?
07:42 MACscr at*
07:44 DV__ joined #gluster
07:48 mreck[nd] joined #gluster
08:02 mreck[nd] Hi, anyone awake ?
08:04 DV joined #gluster
08:11 Slashman joined #gluster
08:18 ivan_rossi joined #gluster
08:19 [Enrico] joined #gluster
08:22 JoeJulian MACscr: Nope, just ubuntu users. Upstart tries to mount before gluster is ready. Either add a sleep to the upstart job, or add more retries and mount servers in fstab.
08:22 JoeJulian mreck[nd]: I was, but I doubt you came to #gluster to learn about my personal life. :D
08:23 MACscr JoeJulian: same issue with debian too
08:24 JoeJulian With systemd?
08:24 MACscr yes
08:24 MACscr Jessie 8.2
08:24 JoeJulian That's odd. I don't have that problem with any of my systemd installs, centos7, fedora 22, archlinux.
08:26 JoeJulian I'm exhausted and heading for bed, but if you're around tomorrow I could probably help take a look. I'm in GMT-8.
08:29 mreck[nd] ahem, no :) short version: After rebuild a failed brick and rebuild gluster in an 3 brick distributed setup any operation is dead slow. The backing devices have normal performance, just one node has changed. Im out of ideas.
08:30 DV__ joined #gluster
08:30 nisroc joined #gluster
08:31 fsimonce joined #gluster
08:35 kshlm joined #gluster
08:37 atalur joined #gluster
08:40 RameshN joined #gluster
08:41 atinm joined #gluster
08:47 kotreshhr joined #gluster
08:48 nishanth joined #gluster
08:49 shubhendu joined #gluster
08:50 ppai joined #gluster
08:50 amye joined #gluster
08:59 gildub joined #gluster
09:01 shubhendu_ joined #gluster
09:05 overclk joined #gluster
09:07 hagarth left #gluster
09:07 hagarth joined #gluster
09:11 Saravana_ joined #gluster
09:12 natarej_ joined #gluster
09:14 nishanth joined #gluster
09:18 timotheus1 joined #gluster
09:19 cjellick joined #gluster
09:23 rjoseph joined #gluster
09:24 DV__ joined #gluster
09:28 kenansulayman joined #gluster
09:36 rjoseph left #gluster
09:39 rjoseph joined #gluster
09:40 bluenemo joined #gluster
09:44 rjoseph joined #gluster
09:45 DV joined #gluster
09:46 MessedUpHare joined #gluster
09:54 rjoseph joined #gluster
09:54 harish_ joined #gluster
09:59 arcolife joined #gluster
10:00 rjoseph joined #gluster
10:03 rjoseph joined #gluster
10:13 DV joined #gluster
10:13 kotreshhr joined #gluster
10:19 cuqa left #gluster
10:22 rjoseph joined #gluster
10:24 nishanth joined #gluster
10:26 kanagaraj joined #gluster
10:26 Slashman joined #gluster
10:28 [Enrico] joined #gluster
10:29 firemanxbr joined #gluster
10:33 Bhaskarakiran_ joined #gluster
10:34 zhangjn joined #gluster
10:36 shubhendu joined #gluster
10:40 zhangjn joined #gluster
10:43 jdarcy joined #gluster
10:46 ctria joined #gluster
10:48 zhangjn joined #gluster
10:50 zhangjn joined #gluster
10:52 vmallika joined #gluster
10:54 Saravana_ joined #gluster
11:08 ramky joined #gluster
11:10 badone joined #gluster
11:12 Bhaskarakiran joined #gluster
11:19 kanagaraj joined #gluster
11:27 yoavz joined #gluster
11:34 kshlm joined #gluster
11:39 rjoseph joined #gluster
11:48 DV__ joined #gluster
11:55 ndevos REMINDER: the Gluster Community Meeting starts in 5 minutes in #gluster-meeting
11:57 ashiq joined #gluster
11:58 kasturi joined #gluster
11:58 sankarshan joined #gluster
11:59 kotreshhr left #gluster
12:01 dusmant joined #gluster
12:04 nishanth joined #gluster
12:05 klaxa joined #gluster
12:08 atalur joined #gluster
12:10 Bhaskarakiran joined #gluster
12:11 MessedUpHare joined #gluster
12:12 overclk joined #gluster
12:22 EinstCrazy joined #gluster
12:26 ashiq_ joined #gluster
12:32 ramteid joined #gluster
12:48 18WABJFVX joined #gluster
12:50 nishanth joined #gluster
12:50 DV__ joined #gluster
12:51 spalai joined #gluster
12:51 spalai left #gluster
12:56 plarsen joined #gluster
12:59 kanagaraj joined #gluster
13:00 lpabon joined #gluster
13:12 klaxa|work joined #gluster
13:16 wnlx joined #gluster
13:22 ndarshan joined #gluster
13:30 d0nn1e joined #gluster
13:35 mhulsman joined #gluster
13:39 atinm joined #gluster
13:45 mhulsman joined #gluster
13:50 hgowtham joined #gluster
13:50 ashiq joined #gluster
13:56 ahino joined #gluster
13:56 Slashman joined #gluster
14:15 ira joined #gluster
14:20 nishanth joined #gluster
14:20 MACscr does gluster use nfs for syncing or the regular client mounting or something? if its not the case, tehn im not using it and how can i disable it?
14:22 kkeithley joined #gluster
14:22 chirino joined #gluster
14:24 zhangjn joined #gluster
14:25 kanagaraj joined #gluster
14:26 ndevos MACscr: for current releases, the Gluster-clients are in charge of the replication/distribution, you should not access the contents on the bricks directly, but have the clients do that for you
14:26 ndevos MACscr: not sure if that answers your question?
14:26 MACscr does not
14:26 MACscr is NFS required
14:27 MACscr why is nfs enabled on my default install
14:27 ndevos no, you do not need to use NFS
14:27 ndevos the Gluster servers provide access to the Gluster volumes over NFS by default
14:27 ndevos you can disable that on a per-volume basis with "gluster volume set <VOLUME> nfs.disable true"
14:30 MACscr ok, cool
14:31 MACscr i sure wish ubuntu 14.04 worked better with it
14:32 MACscr ndevos: how do i know if anything is syncing?
14:33 ndevos MACscr: if you have a replicated volume, the replication in synchronous and you can check on both bricks, new files/dirs should be created instantly there
14:34 MACscr that doesnt really answer my question either =P
14:35 MACscr obviously everything isnt going to be instant as lets say server 2 is down and new files are added to server one. Well obviously when 2 comes back online, it has to sync them to 2
14:35 ndevos MACscr: you can use "gluster volume top ..." to see if clients are writing to the volume, a write involves syncing (all done by the client)
14:35 MACscr and if its 100GB offiles, it might take awhile
14:36 ndevos ah, well, if a server is down, "gluster volume heal ..." can be used to see the status of the files that need healing
14:40 kkeithley joined #gluster
14:42 MACscr that need healing? im assuming its automatically done though, right?
14:43 ndevos yes, there is a self-heal daemon that does the healing at intervals, healing is also checked/performed when something executes the stat() syscall on a file
14:44 nbalacha joined #gluster
14:45 MACscr ndevos: thanks for your time. i appreciate it
14:45 ndevos MACscr: you're welcome, I hope it makes things more understandable
14:47 MACscr i think it does. I pretty much have two lxc guests setup with gluster and a single volume with raw image files in them. These image files are then served as iscsi luns to which they are used as boot file systems for servers
14:47 MACscr fun stuff =P
14:47 MACscr using a bit of keepalived with them as well
14:49 Akee joined #gluster
14:53 DV__ joined #gluster
14:54 hamiller joined #gluster
14:58 ndevos MACscr: oh, yes, sounds interesting, got a blog post about that setup somewhere?
15:01 MACscr when i get it working perfect, i might
15:02 jiffin joined #gluster
15:04 MACscr right  now im having problems with my targets not all populating correctly from its config and gluster not always mounting the client at boot (having to use a upstart script for that since i guess ubuntu trusty doesnt work well with the glusterfs client for regular mounting)
15:07 jiffin joined #gluster
15:09 DV joined #gluster
15:11 jmarley joined #gluster
15:16 dusmant joined #gluster
15:19 DV joined #gluster
15:19 RameshN joined #gluster
15:20 skylar joined #gluster
15:22 jiffin joined #gluster
15:22 ndevos MACscr: if mounting on boot fails, you could try to use autofs or the systemd/automount mount option, some users have success with that
15:25 dgandhi joined #gluster
15:26 dgandhi joined #gluster
15:32 mlncn joined #gluster
15:35 blu_ joined #gluster
15:38 Akee joined #gluster
15:52 ivan_rossi left #gluster
15:56 cjellick joined #gluster
16:02 mlncn joined #gluster
16:05 JoeJulian I still say we need a glusterd command line switch that tells it not to background until all the subservices are started.
16:06 JoeJulian That would cure any startup race condition.
16:11 RameshN joined #gluster
16:11 bennyturns joined #gluster
16:16 csim or systemd integration, with notify (or both)
16:26 wushudoin joined #gluster
16:26 julim joined #gluster
16:27 ahino joined #gluster
16:28 corretico joined #gluster
16:36 tsaavik JoeJulian: doh! thanks for pointing out the timestamp :P
16:37 DV__ joined #gluster
16:38 RameshN joined #gluster
16:38 JoeJulian tsaavik: You're welcome. It's a common misunderstanding.
16:44 jonfatin- left #gluster
16:46 tsaavik was very confusing because someone had written a monitor that called that also. Guess we never got a split-brain before :D
16:51 squizzi_ joined #gluster
16:58 Akee joined #gluster
17:02 DV__ joined #gluster
17:03 timotheus1 joined #gluster
17:05 rafi joined #gluster
17:08 cjellick joined #gluster
17:16 calavera joined #gluster
17:19 jrm16020 joined #gluster
17:28 arcolife joined #gluster
17:29 cjellick joined #gluster
17:29 cjellick left #gluster
17:50 jwd joined #gluster
17:51 jwaibel joined #gluster
17:52 mhulsman joined #gluster
17:54 F2Knight joined #gluster
17:55 bowhunter joined #gluster
17:57 plarsen joined #gluster
17:59 arcolife joined #gluster
18:05 EinstCrazy joined #gluster
18:16 nishanth joined #gluster
18:17 bennyturns joined #gluster
18:26 Rapture joined #gluster
18:35 cliluw joined #gluster
18:41 B21956 joined #gluster
18:56 calavera joined #gluster
18:58 calavera joined #gluster
19:01 calavera joined #gluster
19:20 dgbaley joined #gluster
19:42 jwd joined #gluster
19:53 rafi1 joined #gluster
19:55 lpabon joined #gluster
20:14 mhulsman joined #gluster
20:21 jwaibel joined #gluster
20:31 calavera joined #gluster
20:45 Guest81222 joined #gluster
20:49 MACscr im confused on what bricks are. Isnt a volume what is actually shared? why do i care about bricks? im asking because im trying to setup the proper firewall rules
20:49 MACscr i have two systems and a single volume
20:52 obnox MACscr: a volume consists of (usulally several) bricks
20:53 obnox MACscr: a brick is a local disk (or file system) that is integrated in some way or antother (replication, striping ...) into a cluster-wide gluster volume
20:53 obnox (essentially)
20:59 MACscr obnox: ah, mine are just a directory on each host
21:00 MACscr so i guess i just have two total bricks?
21:02 obnox MACscr: sounds like it.
21:02 obnox MACscr: and then these can be aggregated into the volume.
21:03 obnox MACscr: e.g as replicate volume (corresponding somwhat to the RAID 1 concept)
21:03 obnox (aka mirror)(
21:03 obnox MACscr: or as disctribute volume (corresponding to RAID 0)
21:04 obnox and some more options
21:04 MACscr man, their are way to many firewall rules
21:04 Guest81222 Is there a way within gluster-client to allow for files/directories to be created, but not removed or modified/renamed?
21:04 coreping joined #gluster
21:04 Guest81222 or will I have to utilize something like bindfs to manage this permission model
21:06 obnox Guest81222: like WORM ? http://www.gluster.org/community/documentation/index.php/Features/worm
21:07 Guest81222 yes exactly
21:08 Guest81222 my volume contains scripts that get executed as well. This will allow for execution rights as well?
21:09 Guest81222 i suppose I can easily test that
21:09 Guest81222 thanks!
21:09 obnox Guest81222: welcome
21:10 mhulsman joined #gluster
21:14 jrm16020 joined #gluster
21:30 cliluw joined #gluster
21:31 JoeJulian MACscr: With regard to firewall ,,(ports)
21:31 glusterbot MACscr: glusterd's management port is 24007/tcp (also 24008/tcp if you use rdma). Bricks (glusterfsd) use 49152 & up since 3.4.0 (24009 & up previously). (Deleted volumes do not reset this counter.) Additionally it will listen on 38465-38467/tcp for nfs, also 38468 for NLM since 3.3.0. NFS also depends on rpcbind/portmap on port 111 and 2049 since 3.4.
21:35 jwd joined #gluster
21:50 hamiller joined #gluster
22:32 diegows joined #gluster
22:33 timotheus1 joined #gluster
22:39 F2Knight joined #gluster
22:45 gildub joined #gluster
23:04 zhangjn joined #gluster
23:51 coredump joined #gluster

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