Re: Incorrect return values from glfs_init()

[Anand Avati <anand.avati@xxxxxxxxx>]

ret of 1 is intentional to indicate that glfs_init() could not complete yet. 0 indicates success and you can start issuing fops right away. -1 is definitive failure. When ret is positive, it means initialization could not complete, but glfs_t is still retrying to connect to the server and can succeed in the future (for e.g, if you 'gluster volume start' the volume).

I do understand that it is currently pointless as there is no way to get notified of asynchronous success, but it is part of the asynchronous initialization (glfs_init_async you can see in glfs.c) which will be supported in the future.

Avati

On Wed, Jul 25, 2012 at 10:27 PM, Bharata B Rao <bharata.rao@xxxxxxxxx> wrote:

On Wed, Jul 25, 2012 at 2:08 PM, Bharata B Rao <bharata.rao@xxxxxxxxx> wrote:
> glfs_init() is supposed to return 0  on success and -1 on failure.
>
> When I specify a volume that's not yet "started" from gluster CLI,
> glfs_init() returns 1 with errno 98.

Client volfile
-----------------
volume test-client-0
    type protocol/client
    option remote-host bharata
    option remote-subvolume /test
    option transport-type tcp
end-volume

volume test-dht
    type cluster/distribute
    subvolumes test-client-0
end-volume

volume test-write-behind
    type performance/write-behind
    subvolumes test-dht
end-volume

volume test-read-ahead
    type performance/read-ahead
    subvolumes test-write-behind
end-volume

volume test-io-cache
    type performance/io-cache
    subvolumes test-read-ahead
end-volume

volume test-quick-read
    type performance/quick-read
    subvolumes test-io-cache
end-volume

volume test-md-cache
    type performance/md-cache
    subvolumes test-quick-read
end-volume

volume test
    type debug/io-stats
    option latency-measurement off
    option count-fop-hits off
    subvolumes test-md-cache
end-volume

Client side log
---------------------
[2012-07-26 05:17:57.602104] I [socket.c:3221:socket_init] 0-gfapi:
SSL support is NOT enabled
[2012-07-26 05:17:57.602215] I [socket.c:3236:socket_init] 0-gfapi:
using system polling thread
[2012-07-26 05:17:57.618120] I [socket.c:3221:socket_init]
0-test-client-0: SSL support is NOT enabled
[2012-07-26 05:17:57.618160] I [socket.c:3236:socket_init]
0-test-client-0: using system polling thread
[2012-07-26 05:17:57.618188] I [glfs-master.c:61:notify] 0-gfapi: New
graph 62686172-6174-612d-3930-34332d323031 (0) coming up
[2012-07-26 05:17:57.618206] I [client.c:2141:notify] 0-test-client-0:
parent translators are ready, attempting connect on transport
[2012-07-26 05:17:57.621582] E
[client-handshake.c:1693:client_query_portmap_cbk] 0-test-client-0:
failed to get the port number for remote subvolume
[2012-07-26 05:17:57.621636] W [socket.c:390:__socket_rwv]
0-test-client-0: readv failed (No data available)
[2012-07-26 05:17:57.621662] I [client.c:2089:client_rpc_notify]
0-test-client-0: disconnected
[2012-07-26 05:17:57.621684] I [glfs-master.c:42:glfs_graph_setup]
0-glfs-master: switched to graph 62686172-6174-612d-3930-34332d323031
(0)
[2012-07-26 05:17:57.631304] E [dht-common.c:1372:dht_lookup]
0-test-dht: Failed to get hashed subvol for /
[2012-07-26 05:17:57.631563] E [dht-common.c:1372:dht_lookup]
0-test-dht: Failed to get hashed subvol for /dir1

Server
----------
Not running since volume isn't yet started.

Last glusterd log
(/usr/local/var/log/glusterfs/usr-local-etc-glusterfs-glusterd.vol.log)
------------------------
[2012-07-26 05:17:08.317357] E
[glusterd-store.c:2212:glusterd_store_retrieve_volume] 0-: Unknown
key: brick-0
[2012-07-26 05:17:08.317375] E
[glusterd-store.c:2212:glusterd_store_retrieve_volume] 0-: Unknown
key: brick-1
[2012-07-26 05:17:08.317654] E
[glusterd-store.c:2212:glusterd_store_retrieve_volume] 0-: Unknown
key: brick-0
[2012-07-26 05:17:08.317664] E
[glusterd-store.c:2212:glusterd_store_retrieve_volume] 0-: Unknown
key: brick-1
[2012-07-26 05:17:08.318514] I [glusterd.c:94:glusterd_uuid_init]
0-glusterd: retrieved UUID: b5b47193-5b5e-49da-b28c-8f0aa995e744
[2012-07-26 05:17:08.319091] I
[glusterd-utils.c:1222:glusterd_volume_start_glusterfs] 0-: About to
start glusterfs for brick bharata:/rep1
[2012-07-26 05:17:08.330673] I [socket.c:3221:socket_init]
0-management: SSL support is NOT enabled
[2012-07-26 05:17:08.330726] I [socket.c:3236:socket_init]
0-management: using system polling thread
[2012-07-26 05:17:08.331211] I
[glusterd-utils.c:1222:glusterd_volume_start_glusterfs] 0-: About to
start glusterfs for brick bharata:/rep2
[2012-07-26 05:17:08.339197] I [socket.c:3221:socket_init]
0-management: SSL support is NOT enabled
[2012-07-26 05:17:08.339219] I [socket.c:3236:socket_init]
0-management: using system polling thread
[2012-07-26 05:17:08.339660] I
[glusterd-utils.c:868:glusterd_volume_brickinfo_get] 0-management:
Found brick
[2012-07-26 05:17:08.339899] I
[glusterd-utils.c:868:glusterd_volume_brickinfo_get] 0-management:
Found brick
[2012-07-26 05:17:08.347865] I [socket.c:3221:socket_init]
0-management: SSL support is NOT enabled
[2012-07-26 05:17:08.347895] I [socket.c:3236:socket_init]
0-management: using system polling thread
[2012-07-26 05:17:08.353960] I [socket.c:3221:socket_init]
0-management: SSL support is NOT enabled
[2012-07-26 05:17:08.353983] I [socket.c:3236:socket_init]
0-management: using system polling thread
Given volfile:
+------------------------------------------------------------------------------+
  1: volume management
  2:     type mgmt/glusterd
  3:     option working-directory /var/lib/glusterd
  4:     option transport-type socket,rdma
  5:     option transport.socket.keepalive-time 10
  6:     option transport.socket.keepalive-interval 2
  7:     option transport.socket.read-fail-log off
  8: end-volume

+------------------------------------------------------------------------------+
[2012-07-26 05:17:08.354215] I [socket.c:2081:socket_event_handler]
0-transport: disconnecting now
[2012-07-26 05:17:08.354244] I [socket.c:2081:socket_event_handler]
0-transport: disconnecting now
[2012-07-26 05:17:08.354265] I [socket.c:2081:socket_event_handler]
0-transport: disconnecting now
[2012-07-26 05:17:08.354283] I [socket.c:2081:socket_event_handler]
0-transport: disconnecting now
[2012-07-26 05:17:08.359061] I
[glusterd-pmap.c:237:pmap_registry_bind] 0-pmap: adding brick /rep1 on
port 49154
[2012-07-26 05:17:08.359611] I
[glusterd-pmap.c:237:pmap_registry_bind] 0-pmap: adding brick /rep2 on
port 49155
[2012-07-26 05:17:08.374653] W [socket.c:390:__socket_rwv]
0-socket.management: readv failed (No data available)
[2012-07-26 05:17:08.374715] W [socket.c:390:__socket_rwv]
0-socket.management: readv failed (No data available)
[2012-07-26 05:17:08.376131] W [socket.c:390:__socket_rwv]
0-socket.management: readv failed (No data available)
[2012-07-26 05:17:08.376176] W [socket.c:390:__socket_rwv]
0-socket.management: readv failed (No data available)
[2012-07-26 05:17:14.867131] I
[glusterd-handler.c:852:glusterd_handle_cli_get_volume] 0-glusterd:
Received get vol req
[2012-07-26 05:17:14.867970] I
[glusterd-handler.c:852:glusterd_handle_cli_get_volume] 0-glusterd:
Received get vol req
[2012-07-26 05:17:14.868468] I
[glusterd-handler.c:852:glusterd_handle_cli_get_volume] 0-glusterd:
Received get vol req
[2012-07-26 05:17:14.869070] I
[glusterd-handler.c:852:glusterd_handle_cli_get_volume] 0-glusterd:
Received get vol req
[2012-07-26 05:17:57.621649] W [socket.c:390:__socket_rwv]
0-socket.management: readv failed (No data available)
[2012-07-26 05:17:57.636813] W [socket.c:390:__socket_rwv]
0-socket.management: readv failed (No data available)

_______________________________________________
Gluster-devel mailing list
Gluster-devel@xxxxxxxxxx
https://lists.nongnu.org/mailman/listinfo/gluster-devel