[stgt] tgt memory usage

[Andy Grover]

On 06/19/2013 12:27 AM, Raj Singh wrote:
> tgtadm --lld iscsi --mode target --op new --tid=1 --targetname
> volume-9c2637a7-1c47-41a7-89d3-86dddd8f9082
>
> tgtadm --lld iscsi --mode logicalunit --op new --tid 1 --lun 1 -b
> /dev/cinder-volumes/volume-9c2637a7-1c47-41a7-89d3-86dddd8f9082
>
>
> But when I perform the following command I run out of memory after 3 -
> 4 mins (there are no LUN's or Target's on this machine when I run this
> command):
>
> tgt-admin --update
> iqn.2010-10.org.openstack:volume-9c2637a7-1c47-41a7-89d3-86dddd8f9082
>
>
>
> To fix this for now I upped the memory on my server from 2G to 12G.
> Now when I run the tgt-admin --update command it completes after 8 -
> 10 mins (during this time the memory consumed by tgt-admin goes to
> 10G).
>
> Not sure if this script should take up this much memory for starting a
> single lun and target.

tgt-admin is a hideous, overgrown Perl script, and it sounds like it's 
doing something really inefficient in your case. How big is your config?

One thing you could do is make a copy of the tgt-admin script, and then 
insert print statements to see exactly where things bog down. Is it slow 
before update_targets? Somewhere in process_accounts?

I'd recommend starting at the very bottom, inserting prints in the 
"elsif ($update ne 0) {" block.

Regards -- Andy

--
To unsubscribe from this list: send the line "unsubscribe stgt" in
the body of a message to majordomo at vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html