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; } } }

Restoring iw-store from different TeamSite server

lhdavis

OK, to begin, all servers in question are Windows 2000 Server with SP3 installed; TeamSite 5.5.1 with SP2 installed.

Our desire is to be able to convert one of our Development boxes to Production using a copy of the Production backing store.

First, I obtain a copy of the Production backing store using the following:
1. Execute iwfreeze +1200
2. Use ntbackup to backup the Production backing store,

\iw-store (it's only 1.5 GB)
3. Execute iwfreeze --

Now I've taken this backup of the Production backing store, restored it on my Development server in the

\iw-storeProd directory, and perform the following steps (gleaned from various Knowledgebase articles and DevNet posts):
1. Stop the TeamSite service
2. Backup

\iw-store
3. Rename

\iw-store to

\iw-storeDev
4. Rename

\iw-storeProd to

\iw-store
5. Execute T:\iw-home\bin\iwidmap -u -x C:\store.idmap

\iw-store\default
6. Modify C:\store.idmap to replace all the local users from Production with those on Development (all users exist in both environments so I just have to modify the domain from AAA to BBB for example)
7. Execute T:\iw-home\bin\iwidmap -u -a -I C:\store.idmap

\iw-store\default
8. Reboot server

However, once the server is back up and I log into TeamSite I get the following error in the right side of the GUI:
SCI Error: "No branch found" in pgwriter1.cpp:2684
sci_FindBranchByID(fSCISession, &fServer, archive, branch, &brInfo)
Server version: 5.5.1 Build 6215 Interwoven 20020205

And this is what it said in the iwtrace.log at the time of this error:
TeamSite I18N initialization successful!
[Tue Sep 02 23:47:33 2003]
*** IWServer Locale is --> English_UnitedStates.MS1252@Binary
[Tue Sep 02 23:47:33 2003] [Tue Sep 02 23:47:33 2003] Watch Event Thread started
TeamSite Server Starting.
[Tue Sep 02 23:47:33 2003] iwserver: 5.5.1 Build 8487 Interwoven 20020515
[Tue Sep 02 23:47:33 2003] Reading users and groups...
Local system: "GHCPDS508"
[Tue Sep 02 23:47:35 2003] done. 24 users and 13 groups read.
[Tue Sep 02 23:47:35 2003] Target memory usage: 4075Mb
[Tue Sep 02 23:47:35 2003] Cache size: 30000
[Tue Sep 02 23:47:35 2003] Embedded failsafe is enabled
[Tue Sep 02 23:47:35 2003] Journaling is disabled
[Tue Sep 02 23:47:46 2003] TFSEnvironment: inode cache 30000
[Tue Sep 02 23:47:46 2003]: TNetworkShares::Add Can't NetShareAdd, parameter error 0x0
[Tue Sep 02 23:47:46 2003] Batch Jobs Thread started
[Tue Sep 02 23:47:46 2003] T:\iw-home\etc\iw.cfg: rpc_threadcount=64

I have an eventsubd_out file too with Smiley Surprised

rg.exolab.jms.server.FailedToCreateServerException: JmsServer constructor failure javax.jms.JMSException: Messagegr.initialize error javax.jms.JMSException: Cannot create persistency layer: Cannot acquire lock to the database events
at org.exolab.jms.server.JmsServer.<init>(JmsServer.java:278)
at org.exolab.jms.server.JmsServer.main(JmsServer.java:103)

Can anyone tell me what steps I am either missing or doing wrong?

Find more posts tagged with

Comments

ProfessorX

Hmm sounds like a possible corruption to your store when you backed it up. When I did it I had the same platform but did not have sp3 installed and it worked the way you listed it. But I shut down all IW services instead of running iwfrezze +#### when copying over the iw-store.

Also check this out... and run the ipl command if need be
https://support.interwoven.com/kb/kb_show_article2.asp?ArticleID=2249

Just a suggestion...run some diagnostics against the iw-store like iwfsck

good luck

The Professor-

lhdavis

After using iwfreeze and ntbackup to back-up my backing store, I noticed that when I restored this backing store copy (even on the same box) ntbackup was skipping some files reporting that I had insufficient permissions.

So I had our system administrator restore the backup and this time there were no skipped files. However, when I ran iwfsck -d on the restored copy of iw-store (still on the same server for this test), I saw the following errors:
Command line: T:\iw-home\private\bin\iwfsck -d -b

\restoretest\iw-store\default

***************************************
Walk branches and areas.
***************************************

[Thu Sep 04 05:49:56 2003]
Error, point 0x10c invalid or not found
The main branch 0x10c is missing or invalid.

***************************************
Walk directories, match with areas.
***************************************

[Thu Sep 04 05:49:56 2003]
Error, superArea Point ID 0x100 doesn't exist
[Thu Sep 04 05:49:56 2003]

***************************************
Walk directories and files.
***************************************

Looking up starting vpath /
Unable to locate vpath /

Any suggestions? Is that what I should see before I swap out existing iw-store with the restored iw-store? When I execute iwfsck on my existing iw-store here is what I see:
[Thu Sep 04 05:56:51 2003] Backing store

\iw-store\default is currently locked
by process 1308 on GHCPDS507
Command line: iw-home\private\bin\iwfsck -d
The backing store is locked by another process. For example, the TeamSite server may be running.
Reported results may not be accurate.

***************************************
Walk branches and areas.
***************************************

[Thu Sep 04 05:56:51 2003]

***************************************
Walk directories, match with areas.
***************************************

[Thu Sep 04 05:57:00 2003]
[Thu Sep 04 05:57:53 2003]

***************************************
Walk directories and files.
***************************************

Looking up starting vpath /
[Thu Sep 04 05:57:53 2003]
STATUS: 0 points walked
[Thu Sep 04 05:57:55 2003]
STATUS: 1000 points walked
[Thu Sep 04 05:58:00 2003]
STATUS: 2000 points walked
[Thu Sep 04 05:58:05 2003]
STATUS: 3000 points walked
[Thu Sep 04 05:58:08 2003]
STATUS: 4000 points walked
[Thu Sep 04 05:58:09 2003]
STATUS: 5000 points walked
[Thu Sep 04 05:58:11 2003]
STATUS: 6000 points walked
[Thu Sep 04 05:58:13 2003]
STATUS: 7000 points walked
[Thu Sep 04 05:58:15 2003]
STATUS: 8000 points walked
[Thu Sep 04 05:58:17 2003]
STATUS: 9000 points walked
[Thu Sep 04 05:58:19 2003]
STATUS: 10000 points walked
[Thu Sep 04 05:58:22 2003]
STATUS: 11000 points walked
[Thu Sep 04 05:58:24 2003]
STATUS: 12000 points walked
[Thu Sep 04 05:58:25 2003]
STATUS: 13000 points walked
[Thu Sep 04 05:58:28 2003]
STATUS: 14000 points walked
[Thu Sep 04 05:58:32 2003]
STATUS: 15000 points walked
[Thu Sep 04 05:58:35 2003]
STATUS: 16000 points walked
[Thu Sep 04 05:58:37 2003]
STATUS: 17000 points walked
[Thu Sep 04 05:58:39 2003]
STATUS: 18000 points walked
STATUS: 18823 points walked
Finished walking unwalked points.
[Thu Sep 04 05:58:42 2003]

Total points walked: 344297
Directories walked: 19012
Files walked: 115453
Symlinks walked: 0
Zombies walked: 35952
Total points read from disk: 117417
FINISHED 0 errors found

If I have not replaced the existing iw-store with the restored iw-store and restarted my TeamSite server, should this be the behavior I see using iwfsck? Any thoughts or shared experiences would be most appreciated.

ProfessorX

Sounds like a time to file a support case.

My feelings is that there should be no errors reported when doing any type of iwfsck. Although I currently have an error in our store (support is working on it now with me) TS seems to be running ok.

1 suggestion. If there is not a lot of data in your store copy out the files and recreate a new store (There should be some CLT's on how to do this) then copy those files back.

But once again check with support on the best way to handle this.

The Professor-

lhdavis

Well I took a backup of our Production server's iw-store this time and was able to restore it on one of our development servers. I did run iwfsck after restoring my backup and this time had no errors (which is really a good feeling for those of you who haven't tried it).

Guess the backup copy of the backing store on our other development server must have been corrupt.

Thanks for all the help, Professor!

TJRees

Can anyone confirm if we can perform *incremental* restores of a backing store from one server to another? We intend to do one full *restore* and then incrementally restore daily to make sure that any updates on the old server are transferred over to the new server. Has anyone run into any issues doing this?

Edited by TJRees on 10/06/03 12:30 PM (server time).