Logo 
Search:

Sharepoint Forum

Ask Question   UnAnswered
Home » Forum » Sharepoint       RSS Feeds

Site restore - timeout error

  Asked By: Dana    Date: Mar 27    Category: Sharepoint    Views: 1212

I’ll start with we are using smigrate for this migration since I failed to mention it in the body of my text below.

We have been going through a migration from one set of servers to another set and everything has been going fine thus far. Last weekend we backed up a site that, in compressed format, was a total of 43GB. When restoring the site it skipped over one of its subsites and finished restoring all the rest. I thought maybe it was an error with the servers or the backup file so I backed up just the subsite alone and it was 33GB compressed (the largest up to this point had been 14GB). When attempting to restore the site it came back with a timeout error the first few times. It would just sit there for about 20 minutes and then it would go back to the prompt. I have my report going to a log instead of to the screen and each time it showed only that it had done the manifest file, set up the tasks and then timed out. It never mentions any file transfers or other errors the first few times These servers are live and in use so I thought maybe it was just traffic slowing everything down. Upon checking the CPUs and the Network usage I realized that this was not the issue since each was at below 5% usage. System event logs showed nothing. I checked and found the document which went over the way all the settings should be. Timeout is set for 65000, file upload size is set to 1GB, blocked file types are set the same on both the old and new portal, pagefile is set for a min. of 2GB, etc. None of this helped except that now the error wasn’t a timeout but a “server/site not found. Verify that the name is correct and that the proxy settings are correct”. Name is correct and there is no proxy. On a whim, after the 13th or 14th attempt, I used FrontPage to view the top level site and noticed thet subsite. Took a look at it and there was 461MB (which after ten minutes went up to 514mb even though the restore process hads already errored out) worth of files in there and a total of about 6000 files had been transferred. I deleted the site, recreated, didn’t apply any template (as I had done every time before), verifed via FP that it was empty and tried again. Sure enough it errored out and, when checked via FP, the files were there again and it was exactly the same numbers and it increased in size and files exactly the same way as before. To say I am perplexed is an understatement. It’s stating it cannot find the site yet its still moving data over? This is why I think it is still a timeout issue. I checked the web parts, the files, etc and there seems to be nothing untoward in the old site but to be honest I am not even sure where it is failing since the log file is so. So, as it stands right now I am backing up the site with FP and am going to try with it although I haven’t really had much success with large sites and FP in the past.

It all boils down to, anyone have any ideas on what could be wrong? I have truly run out of ideas and this site is NOT going to be the largest one we migrate. I’d really appreciate any tips. I haven’t mentioned exactly every little detail I checked and set but if it was here:

www.microsoft.com/.../stsf12.mspx

Share: 

 

9 Answers Found

 
Answer #1    Answered By: Jennifer Jones     Answered On: Mar 27

all i can think of is that maybe your sql server's databases could be set  to
full recovery model without regular backups to checkpoint the logs?

 
Answer #2    Answered By: Annie Norris     Answered On: Mar 27

A) I would check with product support services on this. They may be working with some site  size limitations that we don’t know about.

B) I would check to see if it is dying at the same place/same document. Perhaps there is a very large document  that simply won’t move or perhaps the document is an unallowed file  type? (not sure why it would be there in the first place)

C) Run Network Monitor to see what the packets are saying – perhaps there is an error  message in the return packets that don’t appear in the UI.

 
Answer #3    Answered By: Chadd Hahn     Answered On: Mar 27

can/should I ask why you don't backup  the content DB and restore  it to the new SQL/SPS/WSS?

 
Answer #4    Answered By: Cheryl Kelley     Answered On: Mar 27

It’s my understanding that there are more than one site  in the database. Is this correct? He’s trying to move a single site…..

 
Answer #5    Answered By: Sharonda Mcfarland     Answered On: Mar 27

That's correct. Many, many more sites in there. I am just trying to
move the one and its subsites.

As to the other suggestions:

A) I would check with product support services on this. They may
be working with some site  size limitations that we don't know about.

Nothing I can find anywhere except for the 2GB on the manifest  file
and the "recommended" limit of 25GB on the content DBs. We actually
have a call in to PSS at the moment regarding something else so I
will throw this one at them when we next speak.

B) I would check to see if it is dying at the same place/same
document. Perhaps there is a very large document  that simply won't
move or perhaps the document is an unallowed file  type? (not sure why
it would be there in the first place)

I actually have the site owner checking  on which document is the last
one in the transfer before it craps out. As I mentioned earlier, it
IS always at the same point. I really thought  it was an unallowed
file type but I could not find any in the old site.

C) Run Network Monitor to see what the packets are saying –
perhaps there is an error  message in the return packets that don't
appear in the UI.

Hadn't thought of this :)

Since we MUST finish this
migration we are going to do a quick and dirty. Going to use
explorer view to move all the files out of the old site and, once
thats done, back  it up without the docs. We did find one document
folder that contained almost 25GB of data. If this doesn't solve the
problem then I guess we will try the packet sniffing next.

 
Answer #6    Answered By: Damon Garner     Answered On: Mar 27

I suspect a 25GB file  will *always* time  out.

 
Answer #7    Answered By: Christop Mcfadden     Answered On: Mar 27

I hope I didn't give the impression that any one file  was 25GB :P
Actually, even in that case wouldn't Smigrate still break it down into
smaller chunks with the chunk sizes depending on whether you left everything
at default or set  up chunksize using Stsadm?

Anyway, wish me luck. Am in the backup  phase since the files were moved and
will be completing the migration  within the next 36 hours depending on,
well, everything. We just performed a 30GB migration over the weekend with
no problems. I really do believe that it wasn't the size of the site  but
that it was the 1 checked  out file that was locked in read only mode. Guess
we'll never know.

 
Answer #8    Answered By: Victoria Bell     Answered On: Mar 27

As you must already know, smigrate  isn't FAST. You will need most of the
weekend for it to complete

 
Answer #9    Answered By: Cassidy Sharpe     Answered On: Mar 27

Yeah.. We've been averaging about 1.3GB an hour over the last 5 site
migrations, but who's counting..

 
Didn't find what you were looking for? Find more on Site restore - timeout error Or get search suggestion and latest updates.




Tagged: