To help find the bottleneck:

Quickly find network problems by telneting to one of the servers, opening
ftp, ftp "putting" a copy of any convienent 100MB file, then "getting" the
same (copy) of the file back again.  Takes seconds, tells the kbps/mbps
transfer speed between both machines in both directions.  THIS IS IMPORTANT
since I have had machines that can put at 10MBPS, but get at 15KBPS.




-----Original Message-----
From: Steve Rader [mailto:rader AT GINSENG.HEP.WISC DOT EDU]
Sent: Thursday, 29 April, 2004 01:23 PM
To: NETWORKER AT LISTMAIL.TEMPLE DOT EDU
Subject: Re: [Networker] backup mail server very slow (linux)


there's been some interesting talk about where the bottleneck may
be, but there really hasn't been any talk about how to find the
bottleneck...

  try "top", "vmstat 1" to verify that there aren't cpu and/or
  memory bottlenecks

  try "bonnie++" and "time uasm ... > /dev/null" to verify
  there isn't a disk read bottleneck

  try "iperf" to verify there isn't a network througput
  bottleneck

i'd be curious to hear other suggestions from folks on the list
that deal with NetWorker preformance troubleshooting.

iperf is nice because it will show up fast ethernet auto-neg
problems almostly immediately (use -i 1).

if you've got one of those MTAs that delivers each msg into it's own
file, you may never get good throughput.  i routinely see backups of
huge source code trees drop into the ~ 60-80 KB/sec range (whereas
i usually get 2-4 MB/sec) going to DLT4K tapes.

steve
- - -
systems & network guy
high energy physics
university of wisconsin

 > ---- Original Message ----
 > From: David De Maeyer
 > Hi all!
 >
 > We are facing a performance issue when taking a full backup of our mail
 > server (>7000 users, 350 GB of data, running Debian linux).
 >
 > The backup server has no problem with the other linux clients. It is
 > writing at around 100 KB/s... which is far from the 2 to 8 MB/s we are
 > used to have on the other clients.
 >
 > It takes forever to backup all the user directories! And we simply
 > can't find where the bottleneck is... just that there are a lot of I/O
 > on the RAID on which these directories are stored.
 >
 > Would it help to increase/decrease the parallelism for the client
 > and/or other clients (smaller filesystems)?
 >
 > Does somebody on this mailing-list once faced a similar problem? Is the
 > upgrade of the client on the mail server a must?
 >
 > Regards,
 > David
 >
 > server: NetWorker 7.1.1 (running Solaris 8)
 > client: NetWorker 6.1.1 (running Debian linux, IMAP server, 350 GB on
 > external RAID)
 >
 > ___________________________________________________
 > David De Maeyer
 > Roskilde University Center, Department of Computer Science
 > Box 260, Hus 42.1, 4000 Roskilde, Denmark
 > voice (+45) 46 74 38 29 / fax (+45) 46 74 30 72
 >
 > --
 > Note: To sign off this list, send a "signoff networker" command via email
 > to listserv AT listmail.temple DOT edu or visit the list's Web site at
 > http://listmail.temple.edu/archives/networker.html where you can
 > also view and post messages to the list.
 > =*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=

--
Note: To sign off this list, send a "signoff networker" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=

--
Note: To sign off this list, send a "signoff networker" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=