This is a MIME message. If you are reading this text, you may want to
consider changing to a mail reader or gateway that understands how to
properly handle MIME multipart messages.
--=__PartD5F51964.0__=
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
After your next reboot, from Monitor, look at System Resources, LSL
Packet Receive Buffers and enter into the resource tags. Check the
number in use and if (any resource tag actually) the TCPIP.NLN has an
every increasing number in use you may have discovered the issue.
Monitor the count and watch to see if during a backup they (it)
increases to a sum total near or around 10,000 time and again (every 5,
6 or 7 days before a reboot is necessary).
If it does, see Hampus item #3 below.
Scott J.
>>> Marek Gorka <marek.gorka AT sun DOT com> 10/27/04 4:29 AM >>>
thanks for hints ;-)
Hampus Lind wrote:
> Netware is always a problem... :-)
indeed.
[cut]
> 1. We se here that SYS volume takes long time to backup. Therefor we
> have created a policy that only backup SYS one a week and have
excluded
> SYS from the other targets.
we're obliged to backup the SYS daily -> once a week a full backup, the
other days - cumulative. the amount of data to be backed up is approx.
5G,
not too much. when everything is ok with the novell client the full
backup
lasts <0,5 hour, cumulative <20 minutes. the average speed of backup is
about 5MB/s.
as you see, the volume of data is not a problem here. i'm sure that if
there
was 500MB of data to be backed up we'd have the same problem. novell
rox :-)
[cut]
> 3. An upgrade of both the netware platform and netbackup would be
nice
> for you. Netware have improved the TSA function very much since
tsa500.
> We run TSAFS here and it`s much faster. And you with tsafs you can
tune
> your backup env..
unfortunately it's not an option in our customer's environment. the
action
would take 2-3 years to complete :-)
> 4. There is a test untility available at netware download called
> "tsatest". This is a great utility to test your netware env and se
what
> you can expect from your backups.
i'll check it, thanks.
by the way, we are now monitoring via bmc patrol many crucial netware
parameters to pin point the parameter (like buffers or so) which causes
the
novell 'server' to hang. maybe this will drive us to a solution in the
future.
> 5. We dont run OTM here because of all the problems. The OTM cache
file
> probably stays because of that you have to manually kill the job.
> What happens if you dont run OTM? Still same error?
haven't checked that.
> 6. If you set the cache max size to 0 (dynamic) how does it work
then?
> How large does it get?
it varies: sometimes the file grows as much as there's free space,
sometimes
only up to 1,5Gapprox.
> It`s a terrible situation to run Netware and also have to back it
up..
> All software companies say that "we dont priority netware" because no
> one uses it, but we do and we have had so much problem erlier. After
the
> upgrade to netware 6.x and the latest tsa modules things have been
> better...
well. the interesting thing is, that there are two types of novell
netware
in the installation: so called 'NW' servers and 'SD' servers. the same
software, the same configuration, the same platform version. 'NW' run
pretty
ok, 'SD' hang. magic.
regards,
--
Marek Gorka
Client Solutions
Sun Microsystems Poland
_______________________________________________
Veritas-bu maillist - Veritas-bu AT mailman.eng.auburn DOT edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
--=__PartD5F51964.0__=
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; charset=3Diso-8859-1"=
>
<META content=3D"MSHTML 6.00.2800.1400" name=3DGENERATOR></HEAD>
<BODY style=3D"MARGIN: 4px 4px 1px; FONT: 10pt Tahoma">
<DIV>After your next reboot, from Monitor, look at System Resources, LSL =
Packet Receive Buffers and enter into the resource tags. Check the =
number in use and if (any resource tag actually) the TCPIP.NLN has an =
every increasing number in use you may have discovered the issue.</DIV>
<DIV> </DIV>
<DIV>Monitor the count and watch to see if during a backup they =
(it) increases to a sum total near or around 10,000 time and again =
(every 5, 6 or 7 days before a reboot is necessary).</DIV>
<DIV> </DIV>
<DIV>If it does, see Hampus item #3 below.</DIV>
<DIV> </DIV>
<DIV>Scott J.<BR><BR>>>> Marek Gorka <marek.gorka AT sun DOT
com> =
10/27/04 4:29 AM >>><BR></DIV>
<DIV style=3D"COLOR: #000000">thanks for hints ;-)<BR><BR>Hampus Lind =
wrote:<BR><BR>> Netware is always a problem... :-)<BR><BR>indeed.<BR><BR=
>[cut]<BR>> 1. We se here that SYS volume takes long time to backup. =
Therefor we <BR>> have created a policy that only backup SYS one a week =
and have excluded <BR>> SYS from the other targets.<BR><BR>we're =
obliged to backup the SYS daily -> once a week a full backup, the =
<BR>other days - cumulative. the amount of data to be backed up is approx. =
5G, <BR>not too much. when everything is ok with the novell client the =
full backup <BR>lasts <0,5 hour, cumulative <20 minutes. the average =
speed of backup is <BR>about 5MB/s.<BR><BR>as you see, the volume of data =
is not a problem here. i'm sure that if there <BR>was 500MB of data to be =
backed up we'd have the same problem. novell rox :-)<BR><BR>[cut]<BR>> =
3. An upgrade of both the netware platform and netbackup would be nice =
<BR>> for you. Netware have improved the TSA function very much since =
tsa500.<BR>> We run TSAFS here and it`s much faster. And you with tsafs =
you can tune <BR>> your backup env..<BR><BR>unfortunately it's not an =
option in our customer's environment. the action <BR>would take 2-3 years =
to complete :-)<BR><BR>> 4. There is a test untility available at =
netware download called <BR>> "tsatest". This is a great utility to =
test your netware env and se what <BR>> you can expect from your =
backups.<BR><BR>i'll check it, thanks.<BR><BR>by the way, we are now =
monitoring via bmc patrol many crucial netware <BR>parameters to pin point =
the parameter (like buffers or so) which causes the <BR>novell 'server' to =
hang. maybe this will drive us to a solution in the future.<BR><BR>> 5. =
We dont run OTM here because of all the problems. The OTM cache file =
<BR>> probably stays because of that you have to manually kill the =
job.<BR>> What happens if you dont run OTM? Still same error?<BR><BR>hav=
en't checked that.<BR><BR>> 6. If you set the cache max size to 0 =
(dynamic) how does it work then? <BR>> How large does it get?<BR><BR>it =
varies: sometimes the file grows as much as there's free space, sometimes =
<BR>only up to 1,5Gapprox.<BR><BR>> It`s a terrible situation to run =
Netware and also have to back it up.. <BR>> All software companies say =
that "we dont priority netware" because no <BR>> one uses it, but we do =
and we have had so much problem erlier. After the <BR>> upgrade to =
netware 6.x and the latest tsa modules things have been <BR>> better...<=
BR><BR>well. the interesting thing is, that there are two types of novell =
netware <BR>in the installation: so called 'NW' servers and 'SD' servers. =
the same <BR>software, the same configuration, the same platform version. =
'NW' run pretty <BR>ok, 'SD' hang. magic.<BR><BR>regards,<BR><BR>-- =
<BR>Marek Gorka<BR>Client Solutions<BR>Sun Microsystems Poland<BR>_________=
______________________________________<BR>Veritas-bu maillist =
- Veritas-bu AT mailman.eng.auburn DOT edu<BR><A
href=3D"http://mailman.eng.=
auburn.edu/mailman/listinfo/veritas-bu">http://mailman.eng.auburn.edu/mailm=
an/listinfo/veritas-bu</A><BR></DIV></BODY></HTML>
--=__PartD5F51964.0__=--
|
