nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

/iwmnt/default directory not found

abhishek_gupta

Hi,
I had installed TS 5.5.2 on Solaris 8 around 5 months back and was working on it. It was working fine till yesterday. Suddenly I can see that the Servername/default/main cannot be seen in the Workarea in TS GUI. Its blank. The /iwmnt/default folder is corrupted. It shows the icon of a text file. The /local/iw-store is there and I can see the folders inside it, but cant get in. I have seen the log, viz. iwserver,iwtrace and iwevents. But I cant see any latest error there. I was working on some workflows previously and saw those errors. iwreset -a is also happening without fail. Cant open the /etc/mnttab file too.

This used to happen on NT machines with me, where suddenly you find that the Y drive is lost, and when you reboot the machine and restart the server it comes up. But now on Solaris we cant think of rebooting the system.

What could be the problem and how do I resolve it. Need to solve this asap.

Thanks for any help in advance.
-Abhi

Find more posts tagged with

Comments

Adam Stoller

What do you see when you run the following commands:
mount
iwstat -c
iwgethome
iwgetstore
df -k `iwgetstore` `iwgethome`

--fish
(Interwoven Senior Technical Consultant)

abhishek_gupta

Hi,
I think the iwtrace log file shows some errors which might be the cause:
[Fri Jun 13 17:55:56 2003]

Your system is running low on virtual memory (105 Mb is available)
[Fri Jun 13 17:55:56 2003] Server will deactivate all stores if available memory falls below 100 Mb
[Fri Jun 13 17:55:57 2003]
Your system has run low on virtual memory (85 Mb is available)
[Fri Jun 13 17:55:57 2003] All stores are being deactivated
[Fri Jun 13 17:55:57 2003] Memory Monitor: releasing memory...
[Fri Jun 13 17:55:57 2003] Memory Monitor: deactivating stores...
[Fri Jun 13 17:55:57 2003] TEnvironment::Flush flushed 0, unlinked 0
[Fri Jun 13 17:55:57 2003] TEnvironment::Freeze freeze until Sat Jun 12 17:55:57 2004
[Fri Jun 13 17:55:57 2003] TEnvironment::dtor found 0 dirty.
[Fri Jun 13 17:55:57 2003] Deactivated store 0x64 default
[Fri Jun 13 17:55:57 2003] Please shutdown the server.

When I type mount at the prompt I get this:

/ on /dev/vx/dsk/rootvol read/write/setuid/intr/largefiles/onerror=panic/dev=3780000 on Wed Feb 12 12:47:26 2003
/proc on /proc read/write/setuid/dev=3f80000 on Wed Feb 12 12:47:25 2003
/dev/fd on fd read/write/setuid/dev=4080000 on Wed Feb 12 12:47:27 2003
/etc/mnttab on mnttab read/write/setuid/dev=4140000 on Wed Feb 12 12:47:28 2003
/var on /dev/vx/dsk/var read/write/setuid/intr/largefiles/onerror=panic/dev=3780005 on Wed Feb 12 12:47:28 2003
/var/run on swap read/write/setuid/dev=1 on Wed Feb 12 12:47:28 2003
/tmp on swap read/write/setuid/dev=2 on Wed Feb 12 12:47:38 2003
/datavol on /dev/vx/dsk/datadg/vol01 read/write/setuid/intr/largefiles/onerror=panic/dev=3797ed0 on Wed Feb 12 12:47:38 2003
/opt on /dev/vx/dsk/opt read/write/setuid/intr/largefiles/onerror=panic/dev=3780006 on Wed Feb 12 12:47:38 2003
/home on /dev/vx/dsk/home read/write/setuid/intr/largefiles/onerror=panic/dev=3780007 on Wed Feb 12 12:47:38 2003
/iwmnt/default on itiappd2:/iwserver/default remote/read/write/setuid/vers=2/dev=41c004f on Tue May 27 12:40:26 2003
/.iwmnt/default on itiappd2:/iwserver/default remote/read/write/setuid/vers=2/noac/dev=41c0050 on Tue May 27 12:40:26 2003

iwstat -c gives me
Your system has run low on virtual memory, see IWTRACE.LOG for details.
Please shutdown the server.

The iwgethome and iwgetstore show the correct paths and df -k 'iwgetstore' 'iwgethome' shows the following.

/dev/vx/dsk/rootvol 4032142 954657 3037164 24% /
/dev/vx/dsk/opt 6975969 6175327 730883 90% /opt

I guess the problem is that there is not enough memory for teamsite to run. iw-home is in /opt and iw-store is in /local.

Now how do we resolve this problem. The cache settings in iw.cfg are the default ones and I havent changed anything there.

Do help me out asap.

Thanks !
-Abhi

Adam Stoller

Okay - it definitely sounds like the memory situation is the source of your problem.
(1) How much memory do you have on that machine?
(2) What other [enterprise] applications do you have running on that machine?
(3) What service packs and/or patches, if any, do you have applied to TeamSite and TeamSite Templating?
(4) If it gets to be Monday before you answer - or before I can respond - You might want to consider going through Interwoven Support and seeing if they can help guide you through fixing the situation.

--fish
(Interwoven Senior Technical Consultant)

abhishek_gupta

A quick question..will changing the setting of the cachesize in iw.cfg help. Like bringing it down to 20,000. The number of directories and files in the largest branch are not more than 2000 right now. So will bringing the cache size down help?

About the questions which you asked. Its already monday, but still trying to figure out the solution.
1. The RAM is 2 GB and it says /opt (where iw-home) is installed is 90% full, and about 24% for /local.
2. Lots of other small intranet applications are running on the box along with iplanet web server and app server.
3. No service packs and patches are installed.

Thanks,
Abhi

abhishek_gupta

Well, when I did "df -k" , it showed /iwmnt/default and /.iwmnt/default as stale and didnt show any disk space for that. We tried killing manually all the iw processes as they were occupying lots of space, we could see that by typing
"ps -ef|grep iw". Hence killing by kill -9 forced a stop to all these iw processes and freed some space. I think this space is the amount of virtual memory or RAM. We tried restarting the server and then it gracefully started all the processes.
And now df -k shows 24% capacity.

This has solved my problem for the timebieng. But since only 3 MB space is present on the /local and 732KB on the /opt, I have asked my unix admin to allocate around 5 GB space for both /local and /opt, so that there is no problem in future.
Any suggestions/comments?

Thanks,
Abhi

iwovGraduate

Just a comment on how you killed all the iw processes. I would highly recommend using "/etc/init.d/iw.server stop" to stop all the IW processes gracefully instead of kill -9.

abhishek_gupta

I understand that would be the normal way to do it. But the services would't stop by "/etc/init.d/iw.server stop" and had no other option. It used to hang and hence had to go for a forced stop via kill -9.

Thanks!

Adam Stoller

Don't have time right now to try to do a remote diagnosis - if you cannot get help from someone else here - I'd strongly suggest you contact Interwoven Support so that they can work with you more directly.

--fish
(Interwoven Senior Technical Consultant)

Ottawa_IWOV

It is normal during a ps-ef|grep iw to see multiple versions of Interwoven services running (i.e. namely iwwebd?)

Also, is there a way to change the location where the backing store is mounted. We realized after installation that there is not enough space on the current mount.

Any help on these issues would be greatly appreciated.

Thanks.

Lucas Cochrane
lcochrane@deloitte.ca

Adam Stoller

I believe a number of the servicves are handled as master/slave relationships - such that you may see multiple entries for a given process but what you are really seeing is something more like "threads" of the same process. Someone else might be able to provide a more explicit explanation.

I think moving the backing store is essentially a matter of stopping TeamSite; moving the entire backing store to a single partition where you have space; adjusting certain settings in some configuration files (e.g. /etc/defaultiwstore); and restarting TeamSite - though you might want to check with Interwoven Support first.

--fish
Senior Consultant, Quotient Inc.
http://www.quotient-inc.com

Migrateduser

> Also, is there a way to change the location where the backing store is
> mounted. We realized after installation that there is not enough space on
> the current mount.

To change the location of the default backing store (and all subsequently created stores), edit the file /etc/defaultiwstore to the new location. A TeamSite restart will be required for this to become effective. If you are copying a store from one location to another, make sure you preserve file perms and ownership.

If you are using multi-store, there is an /etc/iw.cfg tunable to let you put each store in a different location .. check the manual.

James E. Pace
Quality Engineer
Interwoven