Perl 6 - the future is here, just unevenly distributed

IRC log for #gluster, 2017-09-15

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

All times shown according to UTC.

Time Nick Message
00:11 amye I should do a twitter feed of the found spam poetry
00:12 susant joined #gluster
00:44 aravindavk joined #gluster
00:59 susant joined #gluster
01:01 plarsen joined #gluster
01:05 susant joined #gluster
01:54 ilbot3 joined #gluster
01:54 Topic for #gluster is now Gluster Community - https://www.gluster.org | Documentation - https://gluster.readthedocs.io/en/latest/ | Patches - https://review.gluster.org/ | Developers go to #gluster-dev | Channel Logs - https://botbot.me/freenode/gluster/ & http://irclog.perlgeek.de/gluster/
02:13 kotreshhr joined #gluster
02:26 nh2 joined #gluster
02:35 blu_ joined #gluster
02:44 Wizek_ joined #gluster
03:22 jiffin joined #gluster
03:34 itisravi joined #gluster
03:58 apandey joined #gluster
04:01 kramdoss_ joined #gluster
04:12 Saravanakmr joined #gluster
04:17 nishanth joined #gluster
04:18 susant joined #gluster
04:22 dominicpg joined #gluster
04:29 nbalacha joined #gluster
04:32 ppai joined #gluster
04:46 gyadav joined #gluster
04:47 nbalacha joined #gluster
04:56 skumar joined #gluster
04:58 gyadav_ joined #gluster
04:58 aravindavk joined #gluster
05:04 atinm joined #gluster
05:14 Prasad joined #gluster
05:20 skoduri joined #gluster
05:20 sanoj joined #gluster
05:34 ndarshan joined #gluster
05:36 karthik_us joined #gluster
05:38 nbalacha joined #gluster
05:46 jkroon joined #gluster
05:50 itisravi joined #gluster
05:55 nbalacha joined #gluster
06:01 kramdoss_ joined #gluster
06:02 kdhananjay joined #gluster
06:18 rafi1 joined #gluster
06:21 hgowtham joined #gluster
06:22 skumar_ joined #gluster
06:24 sunnyk joined #gluster
06:24 mbukatov joined #gluster
06:34 ThHirsch joined #gluster
06:34 rastar_dinner joined #gluster
06:36 jtux joined #gluster
06:39 skumar__ joined #gluster
06:51 ivan_rossi joined #gluster
07:21 karthik_us joined #gluster
07:39 poornima_ joined #gluster
07:43 nbalacha joined #gluster
07:49 kramdoss_ joined #gluster
07:50 [diablo] joined #gluster
08:01 _KaszpiR_ joined #gluster
08:20 Bharat_ joined #gluster
08:20 Bharat_ Hi
08:20 glusterbot Bharat_: Despite the fact that friendly greetings are nice, please ask your question. Carefully identify your problem in such a way that when a volunteer has a few minutes, they can offer you a potential solution. These are volunteers, so be patient. Answers may come in a few minutes, or may take hours. If you're still in the channel, someone will eventually offer
08:21 Bharat_ I am having three node gluster cluster for ovirt
08:23 Bharat_ Right now I am getting an error 2017-09-15 08:14:38.050437] C [MSGID: 106003] [glusterd-server-quorum.c:350:glusterd_do_...bricks
08:24 Bharat_ I am newcomer in linux, gluster, ovirt, Kindly help me to fix it
08:48 sanoj joined #gluster
08:49 kharloss joined #gluster
08:58 apandey_ joined #gluster
09:07 apandey__ joined #gluster
09:14 ThHirsch joined #gluster
09:17 buvanesh_kumar joined #gluster
09:20 MrAbaddon joined #gluster
09:22 humblec joined #gluster
09:25 jtux joined #gluster
10:11 buvanesh_kumar joined #gluster
10:16 kshlm joined #gluster
10:31 skoduri joined #gluster
10:33 baber joined #gluster
10:38 bEsTiAn joined #gluster
10:41 jtux joined #gluster
10:50 Wizek_ joined #gluster
10:57 shyam joined #gluster
11:05 karthik_us joined #gluster
11:06 Prasad_ joined #gluster
11:30 Prasad__ joined #gluster
11:33 jiffin joined #gluster
11:34 bluenemo joined #gluster
11:50 freephile joined #gluster
11:55 baber joined #gluster
12:02 davidb2111 left #gluster
12:03 malbx joined #gluster
12:05 theeboat joined #gluster
12:07 theeboat hello, does anybody know why when i share a gluster via VFS I get incredibly slow directory listings vs mounting the volume using FUSE and then sharing the filepath via samba which shows no performance issues?
12:07 theeboat also are there are reasons why i should not do the latter
12:11 plarsen joined #gluster
12:32 nh2 amye: in case you're the right person for this topic too: My mail server sent me: Arrival-Date: 15 Sep 2017 03:46:23 +0200: host mx1.gluster.org[198.61.169.132] said: 451     4.3.5 Server configuration problem (in reply to RCPT TO command)
12:43 bEsTiAn joined #gluster
12:53 jstrunk joined #gluster
12:55 shyam joined #gluster
13:00 shyam joined #gluster
13:03 inodb joined #gluster
13:07 inodb joined #gluster
13:22 inodb joined #gluster
13:26 apandey joined #gluster
13:27 skylar joined #gluster
13:54 juhaj How does glusterfs deal with a situation where I have file A on brick X (no striping) and brick X has 1 GB free, but I want to grow file A by 2 GB: does is move A (or other files from X) to brick Y which has free space, does it start striping A across X and Y, or does it give an error?
14:00 hmamtora joined #gluster
14:00 kkeithley gluster never "starts striping" by itself. FYI, the stripe xlator is deprecated. Nobody should be using it.
14:01 kkeithley Use the sharding option on a volume if you need that functionality.
14:01 kkeithley Independent of that to get striping you have to create the volume that way.
14:06 juhaj Oh, I need to bring myself up to date with gluster, as sharding is something I had not heard of before, but then again it is probably 7 years or so since my last real contact with gluster
14:07 juhaj But what happens in that scenario, "disc full" even though df would imply it is not? (This is what used to happen with lustre in this situation, perhaps still does.)
14:09 arpu joined #gluster
14:15 bEsTiAn joined #gluster
14:19 kpease joined #gluster
14:23 major joined #gluster
14:25 kkeithley that's the same as seven years ago. If you write a file to a distribute volume and the disk fills up, gluster will try to write it on another disk.
14:26 kkeithley If that is successful it'll write a small file with a "link" to where it wrote the file
14:26 vbellur joined #gluster
14:28 kotreshhr joined #gluster
14:35 morse joined #gluster
14:42 jiffin joined #gluster
14:44 [fre] joined #gluster
14:49 farhorizon joined #gluster
15:02 wushudoin joined #gluster
15:03 nadley joined #gluster
15:10 mbrandeis joined #gluster
15:15 nh2 joined #gluster
15:20 shyam joined #gluster
15:22 nbalacha joined #gluster
15:24 inodb joined #gluster
15:26 kpease joined #gluster
15:33 gyadav_ joined #gluster
15:34 juhaj So graceful failover as long as failover is possible at all (i.e. if I have 10 bricks, each with 1 GB free, I still cannot write a 9 GB file)
15:34 juhaj (But if one of them has 2 GB free, I can write a 1.5 GB file regardless of which brick I started with)
15:34 juhaj Is that "link" visible to users?
15:37 kkeithley no. it's probably an xattr
15:40 foster joined #gluster
15:50 juhaj So what is it for? Is it to tell glusterd to go look elsewhere for the data? (So glusterd still believes the file is on the original brick and whatever internal database tells it so is not updated but the xattr is used instead? Just trying to understand what happens under the hood here)
15:51 atrius joined #gluster
15:55 Intensity joined #gluster
15:58 cloph juhaj: whether you can write that depends onwhat type of volume you have, and even if distributed volume, the files are not distributed randomly, but rather a hash is computed that determines on which brick it ends up, so if you barely have the diskspace (some of this might be reserved even), it might not suffice due to more files ending up on a single brick/they're not distributed evenly for that case.
16:00 jkroon joined #gluster
16:01 juhaj cloph: Yes, that's obvious, but in general a distributed volume will rebalance bricks if it can?
16:02 cloph if you mean evenly distribute diskspace usage, then no, that is not guaranteed/how it works.
16:03 cloph It will due to nature of hashes if the number of files is large enough, and distribution of filesize is ~even as well.
16:03 cloph but for only few, large files that  might not be the case.
16:16 ThHirsch joined #gluster
16:18 kkeithley there is no database.  On a distribute (dht) volume the filename is hashed and that determines which brick the file is written to.
16:19 kkeithley If it can't be written there for some reason it is written elsewhere
16:19 juhaj Yea, there's nothing it can do of course if in my 10 bricks, there are 15 files of 1 GB each excpet one which is 200 GB, what I meant the functionality is there and it will do the best it can, but not miracles of course
16:19 kkeithley the "link" helps gluster find the file quickly. In the worst case scenario gluster will look on all the bricks.
16:19 juhaj kkeithley: right
16:19 kramdoss_ joined #gluster
16:19 juhaj thanks
16:26 shaunm joined #gluster
16:32 farhorizon joined #gluster
16:35 zcourts joined #gluster
16:36 zcourts_ joined #gluster
16:36 zcourts joined #gluster
16:39 shyam joined #gluster
16:47 aravindavk joined #gluster
16:53 ivan_rossi left #gluster
17:00 nh2 joined #gluster
17:12 davidb2111 joined #gluster
17:32 anthony25 joined #gluster
17:35 anthony25_ joined #gluster
17:46 atinm joined #gluster
17:49 baber joined #gluster
17:52 MrAbaddon joined #gluster
17:58 prasanth joined #gluster
18:08 WebertRLZ joined #gluster
18:15 shyam joined #gluster
18:16 gbit joined #gluster
18:17 gbit I have two VPS, when I probe from server1 to server2 it connects, server2 says disconnected after probe server1. Is not firewall, I can see them talking on tcpdump too... Wondering what should I try next?
18:17 _KaszpiR_ joined #gluster
18:18 gbit hi all :)
18:23 gbit one thing is both servers are copies, but I have installed gluster after I made the copy..
18:23 gbit so different Uuid's
18:23 baber joined #gluster
18:25 decayofmind joined #gluster
18:26 mbrandeis joined #gluster
18:29 farhorizon joined #gluster
18:29 shyam joined #gluster
18:31 atrius joined #gluster
18:38 gyadav_ joined #gluster
18:45 jiffin joined #gluster
19:04 ogelpre joined #gluster
19:04 rafi joined #gluster
19:07 rafi joined #gluster
19:16 ogelpre i've trouble enabling TLS.
19:16 ogelpre i copied de cert of my CA (KIT-CA) to /etc/ssl/glusterfs.ca, the signed csr to /etc/ssl/glusterfs.pem and the private key to /etc/ssl/glusterfs.pem. i've set the following variables also: client.ssl on, server.ssl on, ssl.certificate-depth 2
19:16 ogelpre But glustershd.log says: "SSL connect error"
19:16 ogelpre Where is my failure?
19:16 ogelpre Which format do i need for the files? All PEM?
19:16 ogelpre i am using version 3.10.5
19:23 leifmadsen_ joined #gluster
19:27 ogelpre after volume stop and start there is a new message: error:1416F086:SSL routines:tls_process_server_certificate:certificate verify failed
19:41 ogelpre ah, i need full certificate chain
19:41 ogelpre i inserted only the certifikate of the ca
19:41 ogelpre i forgot the root ca
19:59 Saravanakmr joined #gluster
20:04 farhorizon joined #gluster
20:20 glusterbot joined #gluster
20:27 ronrib_ joined #gluster
20:41 Humble joined #gluster
21:02 jeffspeff joined #gluster
21:45 major joined #gluster
21:47 malbx joined #gluster
21:47 malbx joined #gluster
21:53 shaunm joined #gluster
21:56 rastar joined #gluster
22:00 snehring joined #gluster
22:11 vbellur joined #gluster
22:33 vbellur joined #gluster
22:35 vbellur joined #gluster
22:35 vbellur joined #gluster
22:50 fury joined #gluster
23:02 malbx joined #gluster
23:16 wushudoin joined #gluster
23:31 baber joined #gluster
23:48 blu_ joined #gluster

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