[Gluster-devel] [Gluster-users] glustershd fail to start after 3.7.7 upgrade

Vijay Bellur vbellur at redhat.com
Tue Feb 2 03:30:13 UTC 2016


On 02/01/2016 10:24 PM, Pranith Kumar Karampuri wrote:
>
>
> On 02/01/2016 10:16 PM, Joe Julian wrote:
>> WTF?
>>
>>         if (!xattrs_list) {
>>                 ret = -EINVAL;
>>                 gf_msg (this->name, GF_LOG_ERROR, -ret,
>> AFR_MSG_NO_CHANGELOG,
>>                         "Unable to fetch afr pending changelogs. Is
>> op-version"
>>                         " >= 30707?");
>>                 goto out;
>>         }
>>
>> This is not going to work. Look at the server upgrade process: Upgrade
>> one side of the replica, wait for self-heals to finish, upgrade the
>> other side. This breaks that process.
>
> Sorry guys, this is my screw up, didn't think of this case in the
> design/review. I will fix this immediately. I hope the rpms are not yet
> available for 3.7.7?
>

Since 3.7.7 is already tagged, let us expedite 3.7.8. It would be good 
to have all the recent fuse memory leak fixes plus a patch for this 
problem in 3.7.8.

-Vijay



More information about the Gluster-devel mailing list