Ceph

10/03/2011

08:18 PM Ceph Revision 5d54398c (ceph): mds: fix possible deadlock in multi-mds setup

This should fix the file_excl case on a file_max update when there
is more than one mds.
If we dont issue caps here ... Brandon Seibel

09/29/2011

11:15 AM CephFS Bug #1538: mds: all clients can and up becoming unresponsive, mds locker waiting for unfreeze

I think I have a fix for this, let me know if it makes sense. Brandon Seibel

09/20/2011

01:43 PM CephFS Bug #1538: mds: all clients can and up becoming unresponsive, mds locker waiting for unfreeze

Narrowed it down to the following scenario:
* At some point earlier on we're imported with a loner
* A client re... Brandon Seibel

09/16/2011

05:59 AM CephFS Bug #1527: mds: assert in handle_cache_expire

I'll try to as well, I'm currently hitting bug 1538 far more often which is making reproducing this difficult. Brandon Seibel

09/13/2011

02:01 PM CephFS Bug #1538 (Resolved): mds: all clients can and up becoming unresponsive, mds locker waiting for unfreeze

Cluster mds config as such: mds e72: 2/2/2 up {0=mds02=up:active,1=mds01=up:active}, 2 up:standby-replay
Tried to ... Brandon Seibel

09/10/2011

06:57 AM CephFS Bug #1527: mds: assert in handle_cache_expire

Unfortunately its not turned up too much but you can see the exporting.
Brandon Seibel

09/09/2011

12:17 PM CephFS Bug #1527: mds: assert in handle_cache_expire

Sorry on second look I'm hitting the probably very rare'ish (and only amplified by my vm test boxes) race in the test... Brandon Seibel

11:33 AM CephFS Bug #1527 (Resolved): mds: assert in handle_cache_expire

Running a multi 4 mds cluster: {0=mds01=up:active,1=mds02=up:active}, 2 up:standby-replay (mds01b and mds02b)
Wh... Brandon Seibel