<div><div>Hi List,</div><div><br></div><div>I do geo-replication on a half a dozen volumes or so across several locations. It works fine except for one, our largest volume, a 2x2 distributed-replicated volume with about 40TB of mixed media on it. All of these servers/volumes are running 3.4.6 on CentOS 6.</div><div><br></div><div>Every few days, geo-replication on this volume will stop, though the status command shows OK.</div><div><br></div><div>I've checked all of the things like clock drift that the geo-rep troubleshooting guide recommends, and everything seems to be fine.</div><div><br></div><div>Hopefully without spamming too much logs, here's the last few lines of what I hope might be relevant whenever logging stops.</div><div><br></div><div>Any ideas would be much apprediated, I've ben running up against this intermittently for months.</div><div><br></div><div>On the master:</div><div><br></div><div>$ cat ssh%3A%2F%2Froot%40192.168.78.91%3Agluster%3A%2F%2F127.0.0.1%3Amedia-vol.log</div><div>[2015-03-19 20:13:15.269819] I [master:669:crawl] _GMaster: completed 60 crawls, 0 turns</div><div>[2015-03-19 20:14:16.218323] I [master:669:crawl] _GMaster: completed 60 crawls, 0 turns</div><div>[2015-03-19 20:15:17.171961] I [master:669:crawl] _GMaster: completed 60 crawls, 0 turns</div><div>[2015-03-19 20:16:18.112601] I [master:669:crawl] _GMaster: completed 60 crawls, 0 turns</div><div>[2015-03-19 20:17:19.52232] I [master:669:crawl] _GMaster: completed 60 crawls, 0 turns</div><div>[2015-03-19 20:18:19.991274] I [master:669:crawl] _GMaster: completed 60 crawls, 0 turns</div><div>[2015-03-19 20:20:06.722600] I [master:669:crawl] _GMaster: completed 23 crawls, 0 turns</div><div>[2015-03-19 20:42:40.970180] I [master:669:crawl] _GMaster: completed 1 crawls, 0 turns</div><div>[2015-03-19 20:47:14.961935] I [master:669:crawl] _GMaster: completed 1 crawls, 0 turns</div><div>[2015-03-19 20:48:22.333839] E [syncdutils:179:log_raise_exception] &lt;top&gt;: connection to peer is broken</div><div><br></div><div><br></div><div>$ cat ssh%3A%2F%2Froot%40192.168.78.91%3Agluster%3A%2F%2F127.0.0.1%3Amedia-vol.gluster.log</div><div>[2015-03-19 20:44:17.172597] W [client-rpc-fops.c:2624:client3_3_lookup_cbk] 0-media-vol-client-2: remote operation failed: Stale file handle. Path: /data/v0.1/images/9c0ea93e-46b2-4a27-b31a-ce26897bd299.jpg (0aadc99c-c3e1-455d-a5ac-4bcf04541482)</div><div>[2015-03-19 20:44:38.314659] W [client-rpc-fops.c:2624:client3_3_lookup_cbk] 0-media-vol-client-2: remote operation failed: Stale file handle. Path: /data/v0.1/images/28ed6b8f-ec29-4066-02be-d7ae9a5a7bb6.jpg (7e97920c-f165-44c3-9868-8dd13cc2b8d0)</div><div>[2015-03-19 20:44:38.314738] W [client-rpc-fops.c:2624:client3_3_lookup_cbk] 0-media-vol-client-3: remote operation failed: Stale file handle. Path: /data/v0.1/images/28ed6b8f-ec29-4066-02be-d7ae9a5a7bb6.jpg (7e97920c-f165-44c3-9868-8dd13cc2b8d0)</div><div>[2015-03-19 20:44:53.029449] W [client-rpc-fops.c:2624:client3_3_lookup_cbk] 0-media-vol-client-2: remote operation failed: Stale file handle. Path: /data/v0.1/images/5d150965-f687-43bb-9gae-8f2d04cb02de.jpg (40676e42-47e0-4c2b-a7b9-9e4101f2e32d)</div><div>[2015-03-19 20:44:53.029557] W [client-rpc-fops.c:2624:client3_3_lookup_cbk] 0-media-vol-client-3: remote operation failed: Stale file handle. Path: /data/v0.1/images/5d150965-f687-43bb-9gae-8f2d04cb02de.jpg (40676e42-47e0-4c2b-a7b9-9e4101f2e32d)</div><div>[2015-03-19 20:45:39.031436] W [client-rpc-fops.c:2624:client3_3_lookup_cbk] 0-media-vol-client-2: remote operation failed: Stale file handle. Path: /data/v0.1/images/38067d80-6a2b-498d-b319-ce6c77354151.jpg (10e53310-0c88-43a7-aa3f-2b48f0720cc7)</div><div>[2015-03-19 20:45:39.031552] W [client-rpc-fops.c:2624:client3_3_lookup_cbk] 0-media-vol-client-3: remote operation failed: Stale file handle. Path: /data/v0.1/images/38067d80-6a2b-498d-b319-ce6c77354151.jpg (10e53310-0c88-43a7-aa3f-2b48f0720cc7)</div><div><br></div><div>On the slave:</div><div><br></div><div>$ cat 8ed3c554-8d5d-4304-9cd1-cb9a17c0fd64:gluster%3A%2F%2F127.0.0.1%3Agtmmedia-storage.gluster.log</div><div>[2015-03-20 01:42:39.715954] I [dht-common.c:1000:dht_lookup_everywhere_done] 0-media-vol-dht: STATUS: hashed_subvol media-vol-replicate-1 cached_subvol null</div><div>[2015-03-20 01:42:39.716792] I [dht-common.c:1000:dht_lookup_everywhere_done] 0-media-vol-dht: STATUS: hashed_subvol media-vol-replicate-1 cached_subvol null</div><div>[2015-03-20 01:42:39.730075] I [dht-common.c:1000:dht_lookup_everywhere_done] 0-media-vol-dht: STATUS: hashed_subvol media-vol-replicate-1 cached_subvol null</div><div>[2015-03-20 01:42:39.730179] I [dht-rename.c:1159:dht_rename] 0-media-vol-dht: renaming /data/v0.1/images-add/.D6C7BEBC-3D2B-413F-96EC-5AE7A44B36C4.jpg.Mmu2MZ (hash=media-vol-replicate-1/cache=media-vol-replicate-1) =&gt; /data/v0.1/images-add/D6C7BEBC-3D2B-413F-96EC-5AE7A44B36C4.jpg (hash=media-vol-replicate-1/cache=&lt;nul&gt;)</div><div>[2015-03-20 01:51:56.502512] I [fuse-bridge.c:4669:fuse_thread_proc] 0-fuse: unmounting /tmp/gsyncd-aux-mount-6Hn48F</div><div>[2015-03-20 01:51:56.596291] W [glusterfsd.c:1002:cleanup_and_exit] (--&gt;/lib64/libc.so.6(clone+0x6d) [0x3455ae88fd] (--&gt;/lib64/libpthread.so.0() [0x3455e079d1] (--&gt;/usr/sbin/glusterfs(glusterfs_sigwaiter+0xcd) [0x40533d]))) 0-: received signum (15), shutting down</div><div>[2015-03-20 01:51:56.596309] I [fuse-bridge.c:5301:fini] 0-fuse: Unmounting '/tmp/gsyncd-aux-mount-6Hn48F'.</div><div><br></div><div>$ cat 8ed3c554-8d5d-4304-9cd1-cb9a17c0fd64:gluster%3A%2F%2F127.0.0.1%3Agtmmedia-storage.log</div><div>[2015-03-16 13:19:54.624100] I [gsyncd(slave):404:main_i] &lt;top&gt;: syncing: gluster://localhost:media-vol</div><div>[2015-03-16 13:19:55.752953] I [resource(slave):483:service_loop] GLUSTER: slave listening</div><div>[2015-03-16 14:02:40.654535] I [repce(slave):78:service_loop] RepceServer: terminating on reaching EOF.</div><div>[2015-03-16 14:02:40.661967] I [syncdutils(slave):148:finalize] &lt;top&gt;: exiting.</div><div>[2015-03-16 14:02:51.987721] I [gsyncd(slave):404:main_i] &lt;top&gt;: syncing: gluster://localhost:media-vol</div><div>[2015-03-16 14:02:53.141150] I [resource(slave):483:service_loop] GLUSTER: slave listening</div><div>[2015-03-17 17:31:25.696300] I [repce(slave):78:service_loop] RepceServer: terminating on reaching EOF.</div><div>[2015-03-17 17:31:25.703775] I [syncdutils(slave):148:finalize] &lt;top&gt;: exiting.</div><div>[2015-03-17 17:31:37.139935] I [gsyncd(slave):404:main_i] &lt;top&gt;: syncing: gluster://localhost:media-vol</div><div>[2015-03-17 17:31:38.228033] I [resource(slave):483:service_loop] GLUSTER: slave listening</div><div>[2015-03-19 20:51:55.965342] I [resource(slave):489:service_loop] GLUSTER: connection inactive for 120 seconds, stopping</div><div>[2015-03-19 20:51:55.979207] I [syncdutils(slave):148:finalize] &lt;top&gt;: exiting.</div></div><div><br></div><div>Thanks,</div><div><br></div><div>-Matt</div>