[Gluster-devel] missing files

David F. Robinson david.robinson at corvidtec.com
Thu Feb 12 16:46:20 UTC 2015


FWIW, starting/stopping a volume that is fast doesn't consistently make 
it slow.  I just tried it again on an older volume... It doesn't make it 
slow.  I also went back and re-ran the test on test3brick and it isn't 
slow any longer.  Maybe there is a time lag after stopping/starting a 
volume before it becomes "fast".

Either way, stopping/starting a "fast" volume only makes it slow for 
some period of time and it doesn't consistently make it slow.  I don't 
think this is the issue.  red-herring.

[root at gfs01bkp /]# gluster volume stop test2brick
Stopping volume will make its data inaccessible. Do you want to 
continue? (y/n) y
[root at gfs01bkp /]# gluster volume start test2brick
volume start: test2brick: success
[root at gfs01bkp /]# mount /test2brick
[root at gfs01bkp /]# cd /test2brick
[root at gfs01bkp test2brick]# time tar -xPf boost_1_57_0.tar; time rm -rf 
boost_1_57_0

real    1m1.124s
user    0m0.432s
sys     0m3.136s

real    0m16.630s
user    0m0.083s
sys     0m0.570s


#... Retest on test3brick after it has been up after a volume restart 
for 20-minutes... Compare this to running the test immediately after a 
restart which gave a time of 3.5-minutes.
[root at gfs01bkp test3brick]#  time tar -xPf boost_1_57_0.tar; time rm -rf 
boost_1_57_0

real    1m17.786s
user    0m0.502s
sys     0m3.278s

real    0m18.103s
user    0m0.101s
sys     0m0.684s



------ Original Message ------
From: "Shyam" <srangana at redhat.com>
To: "David F. Robinson" <david.robinson at corvidtec.com>; "Pranith Kumar 
Karampuri" <pkarampu at redhat.com>; "Justin Clift" <justin at gluster.org>
Cc: "Gluster Devel" <gluster-devel at gluster.org>
Sent: 2/12/2015 11:26:51 AM
Subject: Re: [Gluster-devel] missing files

>On 02/12/2015 11:18 AM, David F. Robinson wrote:
>>Shyam,
>>
>>You asked me to stop/start the slow volume to see if it fixed the 
>>timing
>>issue. I stopped/started homegfs_backup (the production volume with 
>>40+
>>TB) and it didn't make it faster. I didn't stop/start the fast volume
>>to see if it made it slower. I just did that and sent out an email. I
>>saw a similar result as Pranith.
>
>Just to be clear even after restart of the slow volume, we see ~19 
>minutes for the tar to complete, correct?
>
>Versus, on the fast volume it is anywhere between 00:55 - 3:00 minutes, 
>irrespective of start, fresh create, etc. correct?
>
>Shyam



More information about the Gluster-devel mailing list