[Gluster-users] 1.4.0RC6 AFR problems

[Keith Freedman]

At 03:05 PM 12/24/2008, Amar Tumballi (bulde) wrote:
>3) one of the directories had 20GB of data in it.... I went to do an
>ls of the directory and had to wait while it auto-healed all the
>files..  while this is helpful, it would be nice to have gotten back
>the directory listing without having to wait for 20GB of data to get
>sent over the network.
>
>
>Currently this behavior is not going to be changed (at least til 
>1.4.0), because, this can happen only if it is self-healing. And it 
>will make sure things are ok when accessed first time. As it works 
>fine now, we don't want to do a code change upto making a stable release.

I understand the purpose of the functionality and normally would be 
fine with it, but it's just an inconvenient approach.  Ideally, (in 
1.4.1, perhaps), it would return the directory listing to the request 
then do the actual data transfer in the background.  since the 
directory listing doesn't imply that one actually cares about the 
individual file data at this point in time.
Also, if this is the case, then if one of the entries in the 
directory is a directory, does that whole directory get auto-healed 
at the same time, or just files within the current directory?  in 
other words, will this cause an auto-heal of an entire directory 
tree, which would be terribly inconvenient if one has to wait all that time.

>4) while the other server was down, the up server kept failing..
>signal 11?  and I had to constantly remount the filesystem.  It was
>giving me messages about the other node being down which was fine but
>then it'd just die after a while.. consistently.
>
>
>This is fixed in tla, we have made a qa release to internal team, 
>once passes basic tests, will be making next 'RC' release.

I'll do some testing once the next rc is available for download.