E8602 Error when backing up individual mailboxes.

[darrenmills]

Hi,got a problem when backing up individual mailboxes on Exchange 2000.I'm using Arcserve 2000 on Windows NT4.It backs up roughly 50 out of 250 mailboxes we have.The 200 it doesn't backup receive the following error message:E8602 failed to read from database dbname=individual mailbox...... ec=backup agent error --(69)"read/write error in named pipes".Anyone got any clues?

 

[Guest_imported]

Hi,
Could you either apply ARCserve 2000 sp2 + l085115 + the following patch....

http://support.ca.com/Download/patches/asitnt/QO06202.html

Or, apply SP3 and the above patch. I recommend the sp2 solution though.

good luck.

guru - brightstor@hotmail.com

 

[darrenmills]

Thank for you quick reply,i shall give it a go.As i'm speaking with the Arcserve Guru are you aware of the ASMGR.exe Dr.Watson error when modifying jobs.It is really quite laughable the amount of time and effort i have spent with the help or lack of help i should say with CA Support in trying to resolve this problem.Originally i had arcserve on NT4 we then upgraded the server to W2K Server and started getting the above error message.Having spent alot of time talking to CA i have tried various things.Applied SP3,Applied 3 patches one of which refers to the exact problem i get,created new backup jobs,totally reinstalled Arcserve on the same server creating new backup rotation jobs backup jobs etc.... all to no avail.CA suggested reinstalling Arcserve on a totally different server to try and elliminate anything that could be conflicting with it.So i installed a new server with NT4 this time (as this problem only seemed to arise when we upgraded our server to 2000) with nothing else on it other than Arcserve.I applied SP3 as advised by CA and what a suprise i still get this error.I am in the process of trying to get CA to come out and sort this out as i have just lost any patience i ever had with this but am wondering if it is a problem with SP3.Do you know if there is anyway of uninstalling SP3 without reinstalling Arcserve.Backups seem to be okay other than on some server's i'm only getting a backup rate of 20-40mb/min as oppose to 250mb/min on other servers but that's another story.Any help would be greatly appreciated.

Darren

 

[Guest_imported]

Ouch,
Your ARCserve install has some problems. I know this for sure. I have seen your problem before. Not specifically with SP3 though. I have seen it on SP2. Although *ahem* CA recommends SP3, I wouldn't install it just yet. Now in order to roll it back without a re-install, it's going to be quite difficult (if not impossible). However, if we can copy the Database directory under ARCserve as well as the .qsd directory, then after the uninstall and re-install, the transplant of job queue and database may be easy. However any libraries would lose their configuration in this, and would have to be re-configured. Hmmmm, I am trying to think of how we solved the ASMGR.exe problem. It involved a transplant of a previous ASMGR.exe into the ARCserve directory. IF your getting DR.Watsons, is there anychance you could let me know the error? Does this happen whenever modifying the job queue or does it happen when doing whatever? The one I saw happened whenever modifying the job queue. did you have this problem with ARCserve 2000 SP2 at all? Hehehe, I would laugh if it was something silly like not allowing all services to interact with the desktop.

Sorry about all the questions, but let me knoe and I'll see whats going on.

Thanks,

Guru

 

[Guest_imported]

hehehe whooooops,
About the slow backups on some servers. Is your network set to 100Mb Full duplex? Have you tried hard pinging the servers that have slow throughput? Tracert within MS-Dos? Find out where the bottleneck is. In my experience it may simply be a NIC not set to full duplex, or if it is having problems with full duplex, then set it to half for better reliable transmission.
Hang on..... About the ASMGR.exe Dr.Watson, I think I know what it is. Do me a favour and try creating a brand new backup job WITHOUT anything to do with the Exchange agent AT ALL!!! If I am correct, you should have no problem modifying the job. I believe it is actually the badly configured (or un-servicepacked) exchange agent which corrupts the job when you attempt to resubmit it. If this is the case then simply solve the exchange agent problem and we'll have our ASMGR.exe solution. If not then we have some more troubleshooting in store

All the best,

guru - brightstor@hotmail.com

 

[APierce1]

I also receive the Dr. Watson error upon upgrading to ArcServe 2000. My issue turned out to be a corrupt database in the 00000001.QSD folder in the ArcServe directory. Deleting the directory and re-installing SP3 resolved the problem. Tell me, where can I find a list of the issues in SP3?

Best Wishes,
Alan

 

[APierce1]

Also see the MS article Q264731 in regards to MAPI/VAPI issues.

Best wishes,
Alan

 

[darrenmills]

Thanks again for your response,since i'd sent that last message i have been in contact again with CA and they suggested renaming the .qsd folder and then restarting the job engine (thus creating a new .qsd folder) and success, this appears to have resolved the problem.This is weird because this is one of the first things i tried when i had Arcserve installed on W2K Server and it didn't work.To be honest i am just glad it's working now.
The only issue i have now with Arcserve is slow backups with various servers which i will look at next with the help of the advice you have given me.Incidentally on one of the server's i was getting a backup rate of 400mb/min which i thought was fantastic.
Thank's to the Guru with regards to the Individual mailbox error i did what you said and i am now getting succesful backups of individual mailboxes.

 

[Guest_imported]

No probs Darren...

Yeah the .QSD folder. Man, that should of been the first thing CA support suggested. However I have a theory about that. If the Exchange backup was not properly configured, then whenever modifying that job or anything you might get the DR.Watson. But now that the issue with exchange backup was resolved, the ASMGR.exe is not giving Dr.Watsons any more. Could you do me a favour and verify whether or not it was the possible fixing of the corrupt job (i.e. with the bad exchange agent) or if it was solved as soon as you created a new .qsd folder? Just curious for my owning learning experience

guru - brightstor@hotmail.com

 

[APierce1]

I'm finding that the job queue is as fragile as crystal. If I try to modify the top job listed, it always crashes ArcServe Manager, and I must stop the job engine, deleted the job from the .QSD folder, and restart the job engine. Of course, this means I have to re-create the job. ArcServe is starting to leave a bad taste in my mouth.

Best wishes,
Alan Pierce
Ithaca, NY

 

[Guest_imported]

Can you describe to me Alan what kind of job keeps corrupting your queue? Are you backing up clients? Network nodes? Exchange databases? etc....???

Thanks,

guru - brightstor@hotmail.com

 

[Guest_imported]

Actually, APierce1..... Apply this after SP3

http://support.ca.com/Download/patches/asitnt/QO04949.html

guru

 

[tf4]

I too have been getting the infamous E8602 Failed to read from database DBNAME= ... (69) "Read/Write error in named pipes"

I have been trying to troubleshoot this for a couple of weeks now. There seem to be more mailbox folders bieng added each time. When I called Arcserve the technician told me that there is something wrong with my Exchange database and to call Microsoft. I have even run an integrity check on the database and it seems OK. I am sort of scared after they told me there is something wrong with Exchange. I am even having thoughts of re-doing the whole Exchange server. I will try that post SP3 (I do have SP3 installed) patch and try again. Has anyone else had the same error? What seems to be strange is that it's mostly calendars and things that don't get used often. I also have Trend Scanmail running on the Exchange box so I will look into the VAPI problem. Any additional tips would be appreciated...Please tell me that my Exchange server won't eventually die!

 

[ARCserveGuru]

I do not believe that your exchange database has a problem. There may be a conflict with the MAPI/VAPI virus scanning, try setting it to off and trying the back up of the brick level again. Also, apply the

http://support.ca.com/Download/patches/asitnt/QO06202.html

patch after you apply sp3. the patch fixes exchange backup problems with the agent on the brick level, and as such should help solve the e8602 errors.

thanks,

guru

 

[tf4]

That fix worked! I no longer get the error messages in the ArcServe log. I am surprised that CA didn't tell me when I talked to tech support a couple weeks ago.

Maybe you can help me with another problem... On the Exchange server I am constantly getting 2009 errors in the system log. They fill up the entire log and happen basically every minute. Here is the detail:

The server could not expand a table because the table reached the maximum size.

For more information, see Help and Support Center at

http://go.microsoft.com/fwlink/events.asp.

I have tried stopping Trend server protect from scanning the \exchsrvr dir but to no avail. I know that Mailscan runs MAPI only as I have VAPI turned off.

Thank you so much!

 

[ARCserveGuru]

So is trend server actually feeding those errors into the event log? Judging by the error message I don't think it is, but maybe I am wrong. I am not familiar with the trend virus scanning system. This means basically that you have two sets to scanners on your system, one for the file system and one for the database, is this correct?

hehe, call me crazy, but could you have possibly exceeded the table size limits withing exchange on a table or two? More importantly, do you have any size limits set on any mailboxes?

Does anything stop the errors from being created? i.e, have you actually tried stopping the Mailscan antivirus and checked to see if it stops writing to the event log?

regards,

guru

 

[tf4]

I have tried stopping the normal antivirus service but I have not tried stopping the scanmail portion. We still get viruses daily that the scanmail program cleans so I am thinking I will try stopping it this weekend and see if that stops writing to the event log. As for the mailbox limits, I only have warnings when exceeding 80mb and we currently have about 35 mailboxes. Only one is at a high amount 170mb LOL. I would get on him but it is the President of the Co.

I am still trying to contact Trend on this issue since I am pretty sure it is the virus scanner that is causing the event errors. Thank you so much for your help. I love this website MS's knowledgebase doesn't always have all of the answers.

 

[ARCserveGuru]

Hehehe, sometimes even the high and mighty aren't always above the law. Whip this president back in line, hehehe, nah kidding. Any, good luck with trend.

guru

 

[ARCserveGuru]

Hehehe, sometimes even the high and mighty aren't always above the law. Whip this president back in line, hehehe, nah kidding. Anyway, good luck with trend.

guru

 

[tf4]

Trend was no help. They had me turn off all logging. I cleared the event log, all seemed to be fine. I went back to my desk and checked the log...and it is full of the error I may try to get the Pres to archive his mailbox to test these errors. I didn't know that there was a physical limitation in MS Exchange though. I may have to give MS a call on this. I cannot seem to find ANYTHING RE: this problem though!