[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[LDM #TIB-337303]: New IDD relay - ldm is hanging after some time

Hi Pete,

> Thanks for the quick response. You all out at Unidata support are
> awesome, and we really appreciate all you do!

Thanks for the kind words!

> Scientific Linux 6 is a recompile of RedHat Enterprise 6, which is
> sourced in part from Fedora 12, 13, and 14. From the wikipedia page on
> RedHat Enterprise:

OK, interesting...

> What file system type are you using for your queue file system?

A quick check of one of the idd.unidata.ucar.edu cluster nodes (one
with a 20 GB LDM queue) seems to indicate we are using ext4:

% cat /etc/fstab

# /etc/fstab
# Created by anaconda on Sat Mar  6 06:31:55 2010
# Accessible filesystems, by reference, are maintained under '/dev/disk'
# See man pages fstab(5), findfs(8), mount(8) and/or blkid(8) for more info
UUID=74a809ff-8c93-4791-bb23-4dcaa91ddeaa /                       ext4    
defaults        1 1
UUID=4aa50e94-3e1f-4dd5-910c-6fd1f7078063 swap                    swap    
defaults        0 0
tmpfs                   /dev/shm                tmpfs   defaults        0 0
devpts                  /dev/pts                devpts  gid=5,mode=620  0 0
sysfs                   /sys                    sysfs   defaults        0 0
proc                    /proc                   proc    defaults        0 0

Way back when I was investigating the problem of putting an LDM queue on a
RAID under Fedora 1 or 3, I tried a variety of file system types, and got
no clear indication that one was any better than the other.  When the queue
was on a non-RAID ext2 file system, however, things worked very nicely.
I have to say that my experience is now very old, and the fact that we
are running idd.unidata.ucar.edu cluster nodes with large LDM queues (20 GB)
on RAIDs currently indicates that things have changed over the years.
Mike Schmidt is the best one to answer specific questions on the RAID
setup on our cluster nodes.  I will touch base with him tomorrow to see
if he has some pearls of wisdom to throw in (he is out of the office at
the moment).

> I've got
> it set up as an xfs right now, but have the option of using ext2/3/4
> instead. Any experience what works best for a file system holding an ldm
> queue?

The /etc/fstab listing above seems to indicate we are using ext4.

> I'll get back to you and let you know if the small queue hangs or not.
> Should be soon.

-- Later --

> With a 1 Gb queue, the ldm did not hang after more than an hour. I'm
> going to set it up now as a non-raid system, and try it with varying
> sizes of queue to see what happens. Again, I'll keep you posted.

Interesting result!  We will be waiting eagerly for further info...


Unidata User Support                                    UCAR Unidata Program
(303) 497-8642                                                 P.O. Box 3000
address@hidden                                   Boulder, CO 80307
Unidata HomePage                       http://www.unidata.ucar.edu

Ticket Details
Ticket ID: TIB-337303
Department: Support LDM
Priority: Normal
Status: Closed

NOTE: All email exchanges with Unidata User Support are recorded in the Unidata inquiry tracking system and then made publicly available through the web. If you do not want to have your interactions made available in this way, you must let us know in each email you send to us.