bugGluster - Bugs: bug #25180, gluster continuously attempts to...

 
 

bug #25180: gluster continuously attempts to reconnect to volumes for which it is already connected

Submitted by:  swankier <swankier>
Submitted on:  Fri 26 Dec 2008 11:43:34 PM UTC  
 
Category: GlusterFSSeverity: 3 - Normal
Priority: 5 - NormalItem Group: Improper behaviour
Status: InvalidPrivacy: Public
Assigned to: NoneOriginator Name: -
Open/Closed: ClosedRelease: 
Operating System: GNU/LinuxReproducibility: Every Time

Add a New Comment(Rich Markup)
   

You are not logged in

Please log in, so followups can be emailed to you.

 

(Jump to the original submission Jump to the original submission)

Tue 23 Jun 2009 06:09:25 PM UTC, comment #7:

Debug log messages are printed during the connection initialization duration and should be ignored, as they are harmless.

--
Gowda

Basavanagowda Kanur <gowda>
Project Member
Mon 05 Jan 2009 10:00:33 AM UTC, comment #6:

my log files have rotated away (well, i deleted them all before trying the latest TLA.)

I guess I'll have to keep my eye on it from now on.

swankier <swankier>
Mon 05 Jan 2009 09:57:49 AM UTC, comment #5:

swankier,
Did you observe any call_bail messages just before these logs? If so this is normal because, call_bail actually disconnects the transport and a new connection establishment is tried.

Raghavendra <raghavendra>
Project Member
Mon 05 Jan 2009 09:43:19 AM UTC, comment #4:

raghavendra,

This happens to a mounted system... after having been mounted for a long period of time, and while it's being used.

This is normal too?

In what circumstance would this not be normal for you?

swankier <swankier>
Mon 05 Jan 2009 09:35:15 AM UTC, comment #3:

After the initial connection between server and client transports, glusterfs needs for a protocol level handshake b/w client and server succeed. This handshake checks for parameters like whether client can be authenticated, proper version number of client etc., The log messages you observed are logged by the client in the time window between the initial connection establishment b/w transports (socket/tcp in your case) and before the response for the handshake is received from server. After the response for the handshake is got, these messages will not be logged. Hence there is no major performance problem over here.

Raghavendra <raghavendra>
Project Member
Mon 05 Jan 2009 08:42:00 AM UTC, comment #2:

even if it is harmless,

is this not a performance problem?

why is this code path being ran?

swankier <swankier>
Mon 29 Dec 2008 08:59:36 PM UTC, comment #1:

this seems to occur anytime the following function is called:

server-protocol.c:3288:server_lookup_resume

swankier <swankier>
Fri 26 Dec 2008 11:43:34 PM UTC, original submission:

running the client in debug mode, I see the following:

2008-12-26 18:34:57 D [client-protocol.c:5690:client_protocol_reconnect] nas2: attempting reconnect
2008-12-26 18:34:57 D [socket.c:924:socket_connect] nas2: connect () called on transport already connected
2008-12-26 18:35:07 D [client-protocol.c:5690:client_protocol_reconnect] nas2: attempting reconnect
2008-12-26 18:35:07 D [socket.c:924:socket_connect] nas2: connect () called on transport already connected

===========

web1:/www/photo_rolls_ite/it# cat /usr/local/etc/glusterfs/glusterfs-client.vol
volume nas1brick1
type protocol/client
option transport-type tcp/client
option remote-host 10.5.1.180
option remote-subvolume nas1brick1p
end-volume

volume nas1brick2
type protocol/client
option transport-type tcp/client
option remote-host 10.5.1.180
option remote-subvolume nas1brick2p
end-volume

volume nas1-ns
type protocol/client
option transport-type tcp/client
option remote-host 10.5.1.180
option remote-subvolume nas1-nsp
end-volume

volume nas1
type cluster/unify
option namespace nas1-ns
option scheduler rr
subvolumes nas1brick1 nas1brick2
end-volume

volume nas2
type protocol/client
option transport-type tcp/client
option remote-host 10.10.1.185
option remote-subvolume nas2
end-volume

volume array
type cluster/afr
subvolumes nas2 nas1
option favorite-child nas2
end-volume

volume writebehind
type performance/write-behind
option aggregate-size 1MB
option window-size 3MB
option flush-behind on
subvolumes array
end-volume

swankier <swankier>

 

(Note: upload size limit is set to 16384 kB, after insertion of the required escape characters.)

Attach File(s):
   
   
Comment:
   

No files currently attached

 

Depends on the following items: None found

Items that depend on this one: None found

 

Carbon-Copy List
  • -unavailable- added by gowda (Posted a comment)
  • -unavailable- added by raghavendra (Posted a comment)
  • -unavailable- added by swankier (Submitted the item)
  •  

    Do you think this task is very important?
    If so, you can click here to add your encouragement to it.
    This task has 0 encouragements so far.

    Only logged-in users can vote.

     

    Please enter the title of George Orwell's famous dystopian book (it's a date):

     

     

    Follow 3 latest changes.

    Date Changed By Updated Field Previous Value => Replaced By
    Tue 23 Jun 2009 06:09:25 PM UTCgowdaOpen/ClosedOpen=>Closed
    Mon 05 Jan 2009 09:35:47 AM UTCraghavendraStatusNone=>Invalid
    Mon 05 Jan 2009 09:35:15 AM UTCraghavendraOriginator Name=>-

    Back to the top


    Powered by Savane 3.1-cleanup1