Jerome,
I believe first and second behavior is same as described in
https://bugzilla.redhat.com/show_bug.cgi?id=820821 by Andrew. I'm not
yet entirely sure WHY is happening.
Third one, flushing, is very important. Without flush, buffer may start
to overload and it causes really bad behavior (there was BZ with this
problem).
I would like Steve to review your patch, but for me it looks like ok.
Regards,
Honza
Jerome FLESCH napsal(a):
Hello,
When upgrading from Corosync 1.2.8 to Corosync 1.4.2/1.4.3, some nasty bugs appeared on our clusters. I observed the following bad behaviors:
1) A process connected to Corosync with CPG wasn't correctly informed that there are other processes connected on other processors. It also didn't get their messages
2) A process sending messages with CPG never received copies of its messages
3) 1 ring out of 2 went up/down quite often
The behaviors 1 and 2 are very hard for us to reproduce, but we are able to get the behavior 3 quite easily.
The simplest setup we found to get it is the following:
- 2 VirtualBox VMs, connected by 2 network interfaces (vboxnet0, vboxnet1 ; one for each ring)
- OS: Linux (Debian stable)
- On one of the VMs, a test program sending some CPG messages (see the script "test_corosync.sh" joined to this mail for example)
Here are the Corosync logs we get when we do this setup:
Jun 06 16:23:40 corosync [TOTEM ] A processor joined or left the membership and a new membership was formed.
Jun 06 16:23:40 corosync [CPG ] chosen downlist: sender r(0) ip(192.168.56.104) r(1) ip(192.168.57.104) ; members(old:1 left:0)
Jun 06 16:23:40 corosync [MAIN ] Completed service synchronization, ready to provide service.
Jun 06 16:24:37 corosync [TOTEM ] Marking ringid 1 interface 192.168.57.105 FAULTY
Jun 06 16:24:38 corosync [TOTEM ] Automatically recovered ring 1
Jun 06 16:25:33 corosync [TOTEM ] Marking ringid 1 interface 192.168.57.105 FAULTY
Jun 06 16:25:34 corosync [TOTEM ] Automatically recovered ring 1
Jun 06 16:26:35 corosync [TOTEM ] Marking ringid 1 interface 192.168.57.105 FAULTY
Jun 06 16:26:36 corosync [TOTEM ] Automatically recovered ring 1
(...)
The second ring goes down about every 2 minutes and automatically back up right after.
We spent some times looking for the commit that introduced this bug, and it appears it's due the following one:
Corosync 1.3.3 -> 1.3.4: e27a58d93d0d3795beb550f87b660c9c04f11386
Corosync 1.4.1 -> 1.4.2: be608c050247e5f9c8266b8a0f9803cc0a3dc881
Commit message: Ignore memb_join messages during flush operations
I had a look at this commit, and it seems to me it's dropping too many packets:
Because of this commit, while totemrrp_recv_flush() is called, Corosync drops memb_join packets, but also ORF tokens. In the end, it seems that sometimes, we drop so many of them that Corosync marks the ring as faulty.
To fix that, I've made the patch joined to this mail (corosync-fix-token-drop.patch).
However I wonder why this packet dropping is done at such a low layer. Wouldn't it be more appropriate to do it in totemsrp.c ?
Moreover, it seems to me that totemrrp_recv_flush() is called every times Corosync get an ORF token (in message_handler_orf_token()). It seems weird to me because the commit message says the packets should only be dropped when we are in gather state to avoid switching suddenly to recovery state.
Also, could you tell me if this packet dropping could explain the 2 other behaviors I observed ?
Thanks in advance,
Regards,
_______________________________________________
discuss mailing list
discuss@xxxxxxxxxxxx
http://lists.corosync.org/mailman/listinfo/discuss
_______________________________________________
discuss mailing list
discuss@xxxxxxxxxxxx
http://lists.corosync.org/mailman/listinfo/discuss
