[Gluster-devel] test failure - tar on gluster

Anand Avati avati at zresearch.com
Sun Jun 24 19:51:35 UTC 2007


Harris,
  please refer -

http://www.gluster.org/docs/index.php/GlusterFS_FAQ#How_do_I_add_a_new_node_to_an_already_running_cluster_of_GlusterFS

you can extend that idea. if you have N storage nodes, take the
skeleton.cpio from N-1 and 'add' (extract) them to the N'th node, build a
fresh skeleton from the N'th node and 'add' (extract) them to the reamining
N-1 nodes. (this is crudely the logic how self-heal would work in unify)

thanks,
avati


2007/6/25, Harris Landgarten <harrisl at lhjonline.com>:
>
> 2.4 I am trying to make sure data is consistent and not the cause of
> namespace volume problems on 2.5
> Harris
> ----- Original Message -----
> From: "Anand Avati" <avati at zresearch.com>
> To: "Harris Landgarten" <harrisl at lhjonline.com>
> Cc: "gluster-devel" <gluster-devel at nongnu.org>
> Sent: Sunday, June 24, 2007 3:41:04 PM (GMT-0500) America/New_York
> Subject: Re: [Gluster-devel] test failure - tar on gluster
>
> Is your previous mail in refernce to glusterfs--mainline--2.4 or
> glusterfs--mainline--2.5? (my previous mail was in reference to
> glusterfs--mainline--2.4)
>
> thanks,
> avati
>
>
> 2007/6/25 , Harris Landgarten < harrisl at lhjonline.com >:
>
> I will try that. Somehow, some skel inconsistencies have gotten into the
> /mnt/glusterfs/backups tree. The bricks have differing numbers of
> directories. How can I fix this?
>
> Harris
>
> ----- Original Message -----
> From: "Anand Avati" < avati at zresearch.com >
> To: "Harris Landgarten" < harrisl at lhjonline.com >
> Cc: "gluster-devel" < gluster-devel at nongnu.org >
> Sent: Sunday, June 24, 2007 3:29:44 PM (GMT-0500) America/New_York
> Subject: Re: [Gluster-devel] test failure - tar on gluster
>
> Harris,
> from the glusterfs--mainline--2.4 tree, you can add a 'option
> transport-timeout <big>' in the protocol/client volume where <big> is in
> seconds (say 600). also if you are using write-behind xlator, adding 'option
> flush-behind off' into that volume will also help in eliminating this error.
> could you see if those helped you out?
>
> thanks
> avati
>
>
> 2007/6/15 , Harris Landgarten < harrisl at lhjonline.com >:
>
> Using mainline-2.4-patch-181
>
> I tried the following
>
> cd /mnt/glusterfs
> sudo tar -cvjf test.tbz secondary/0/9
>
> secondary/0/9 contains 62104 files ranging from 100b to 30mb in size
>
> The idea of the test was to force reading and writing from the same
> gluster client for an extented period.
>
> The test failed with the following from client and brick logs.
>
> client:
>
> [ Jun 15 08:31:42 ] [DEBUG/tcp-client.c:54/do_handshake()]
> transport/tcp-client:dictionary length = 50
> [ Jun 15 08:42:39 ] [CRITICAL/client- protocol.c:218/call_bail()]
> client/protocol:bailing transport
> [ Jun 15 08:42:39 ] [DEBUG/tcp.c:123/cont_hand()] tcp:forcing
> poll/read/write to break on blocked socket (if any)
> [ Jun 15 08:42:39 ] [CRITICAL/client-protocol.c :218/call_bail()]
> client/protocol:bailing transport
> [ Jun 15 08:42:39 ] [DEBUG/tcp.c:123/cont_hand()] tcp:forcing
> poll/read/write to break on blocked socket (if any)
> [ Jun 15 08:42:39 ] [CRITICAL/client-protocol.c:218/call_bail()]
> client/protocol:bailing transport
> [ Jun 15 08:42:39 ] [DEBUG/tcp.c:123/cont_hand()] tcp:forcing
> poll/read/write to break on blocked socket (if any)
> [ Jun 15 08:42:39 ] [CRITICAL/client-protocol.c:218/call_bail()]
> client/protocol:bailing transport
> [ Jun 15 08:42:39 ] [DEBUG/tcp.c:123/cont_hand()] tcp:forcing
> poll/read/write to break on blocked socket (if any)
> [ Jun 15 08:42:39 ] [CRITICAL/client-protocol.c:218/call_bail()]
> client/protocol:bailing transport
> [ Jun 15 08:42:39 ] [DEBUG/tcp.c:123/cont_hand()] tcp:forcing
> poll/read/write to break on blocked socket (if any)
> [ Jun 15 08:42:39 ] [CRITICAL/client- protocol.c:218/call_bail()]
> client/protocol:bailing transport
> [ Jun 15 08:42:39 ] [DEBUG/tcp.c:123/cont_hand()] tcp:forcing
> poll/read/write to break on blocked socket (if any)
> [ Jun 15 08:42:39 ] [CRITICAL/client-protocol.c :218/call_bail()]
> client/protocol:bailing transport
> [ Jun 15 08:42:39 ] [DEBUG/tcp.c:123/cont_hand()] tcp:forcing
> poll/read/write to break on blocked socket (if any)
> [ Jun 15 08:42:39 ] [CRITICAL/client- protocol.c:218/call_bail()]
> client/protocol:bailing transport
> [ Jun 15 08:42:39 ] [DEBUG/tcp.c:123/cont_hand()] tcp:forcing
> poll/read/write to break on blocked socket (if any)
> [ Jun 15 08:42:39 ] [CRITICAL/client-protocol.c :218/call_bail()]
> client/protocol:bailing transport
> [ Jun 15 08:42:39 ] [DEBUG/tcp.c:123/cont_hand()] tcp:forcing
> poll/read/write to break on blocked socket (if any)
> [ Jun 15 08:42:39 ] [CRITICAL/client-protocol.c:218/call_bail()]
> client/protocol:bailing transport
> [ Jun 15 08:42:39 ] [DEBUG/tcp.c:123/cont_hand()] tcp:forcing
> poll/read/write to break on blocked socket (if any)
> [ Jun 15 08:42:39 ] [CRITICAL/client- protocol.c:218/call_bail()]
> client/protocol:bailing transport
> [ Jun 15 08:42:39 ] [DEBUG/tcp.c:123/cont_hand()] tcp:forcing
> poll/read/write to break on blocked socket (if any)
> [ Jun 15 08:42:39 ] [CRITICAL/client-protocol.c :218/call_bail()]
> client/protocol:bailing transport
> [ Jun 15 08:42:39 ] [DEBUG/tcp.c:123/cont_hand()] tcp:forcing
> poll/read/write to break on blocked socket (if any)
> [ Jun 15 08:42:39 ] [CRITICAL/client-protocol.c:218/call_bail()]
> client/protocol:bailing transport
> [ Jun 15 08:42:39 ] [DEBUG/tcp.c:123/cont_hand()] tcp:forcing
> poll/read/write to break on blocked socket (if any)
> [ Jun 15 08:42:39 ] [CRITICAL/client-protocol.c:218/call_bail()]
> client/protocol:bailing transport
> [ Jun 15 08:42:39 ] [DEBUG/tcp.c:123/cont_hand()] tcp:forcing
> poll/read/write to break on blocked socket (if any)
> [ Jun 15 08:42:39 ] [CRITICAL/client- protocol.c:218/call_bail()]
> client/protocol:bailing transport
> [ Jun 15 08:42:39 ] [DEBUG/tcp.c:123/cont_hand()] tcp:forcing
> poll/read/write to break on blocked socket (if any)
> [ Jun 15 08:42:39 ] [CRITICAL/client- protocol.c :218/call_bail()]
> client/protocol:bailing transport
> [ Jun 15 08:42:39 ] [DEBUG/tcp.c:123/cont_hand()] tcp:forcing
> poll/read/write to break on blocked socket (if any)
> [ Jun 15 08:42:39 ] [CRITICAL/client-protocol.c :218/call_bail()]
> client/protocol:bailing transport
> [ Jun 15 08:42:39 ] [DEBUG/tcp.c:123/cont_hand()] tcp:forcing
> poll/read/write to break on blocked socket (if any)
> [ Jun 15 08:42:39 ] [CRITICAL/client-protocol.c:218/call_bail()]
> client/protocol:bailing transport
> [ Jun 15 08:42:39 ] [DEBUG/tcp.c:123/cont_hand()] tcp:forcing
> poll/read/write to break on blocked socket (if any)
> [ Jun 15 08:42:39 ] [ERROR/common-utils.c:110/full_rwv()]
> libglusterfs:full_rwv: 91574 bytes r/w instead of 131418 (Broken pipe)
> [ Jun 15 08:42:39 ] [ERROR/client-protocol.c:204/client_protocol_xfer()]
> protocol/client:transport_submit failed
> [ Jun 15 08:42:39 ] [DEBUG/client-protocol.c:2609/client_protocol_cleanup()]
> protocol/client:cleaning up state in transport object 0x80773e8
> [ Jun 15 08:42:39 ] [DEBUG/client-protocol.c:2712/client_protocol_interpret()]
> protocol/client:frame not found for blk with callid: 62211
> [ Jun 15 08:42:39 ] [DEBUG/client-protocol.c:2609/client_protocol_cleanup()]
> protocol/client:cleaning up state in transport object 0x80773e8
> [ Jun 15 08:42:39 ] [CRITICAL/tcp.c:81/tcp_disconnect()]
> transport/tcp:client1: connection to server disconnected
>
>
> Brick1:
> [ Jun 15 08:42:39 ] [ERROR/common-utils.c:110/full_rwv()]
> libglusterfs:full_rwv: 28656 bytes r/w instead of 65744 (Connection reset by
> peer)
> [ Jun 15 08:42:39 ] [ERROR/proto-srv.c:117/generic_reply()]
> protocol/server:transport_writev failed
> [ Jun 15 08:42:39 ] [ERROR/tcp.c:110/tcp_except()] transport/tcp:shutdown
> () - error: Transport endpoint is not connected
> [ Jun 15 08:42:39 ] [ERROR/common-utils.c:110/full_rwv()]
> libglusterfs:full_rwv: 0 bytes r/w instead of 65746 (Broken pipe)
> [ Jun 15 08:42:39 ] [ERROR/proto-srv.c:117/generic_reply()]
> protocol/server:transport_writev failed
> [ Jun 15 08:42:39 ] [ERROR/tcp.c:110/tcp_except()] transport/tcp:shutdown
> () - error: Transport endpoint is not connected
> [ Jun 15 08:42:39 ] [DEBUG/proto-srv.c:2826/open_file_cleanup_fn()]
> protocol/server:force releaseing file 0x8053e90
> [ Jun 15 08:42:39 ] [DEBUG/proto-srv.c:2826/open_file_cleanup_fn()]
> protocol/server:force releaseing file 0x8053068
> [ Jun 15 08:42:39 ] [DEBUG/proto-srv.c:2867/proto_srv_cleanup()]
> protocol/server:cleaned up xl_private of 0x8052f98
> [ Jun 15 08:42:39 ] [CRITICAL/tcp.c:81/tcp_disconnect()]
> transport/tcp:server: connection to server disconnected
> [ Jun 15 08:42:39 ] [DEBUG/tcp-server.c:229/gf_transport_fini()]
> tcp/server:destroying transport object for 10.255.62.81:1022 (fd=7)
>
> Brick2:
>
> no errors
>
>
>
>
> _______________________________________________
> Gluster-devel mailing list
> Gluster-devel at nongnu.org
> http://lists.nongnu.org/mailman/listinfo/gluster-devel
>
>
>
> --
> Anand V. Avati
>
>
>
> --
> Anand V. Avati
>



-- 
Anand V. Avati



More information about the Gluster-devel mailing list