dbcheck just won't work :( 1

[segil]

Hi,

I'm afraid my Arcserve database is corrupt. I use Brightstor v11.1. I've tried to run dbcheck to test the database, but it won't authenticate me. I've searched this site extensively, and all I can find is that you should use the following syntax:

dbcheck -a -L rdsnp;admin;secret [db name]

But EVERYTIME I get the error message
"failed to log onto server 'rdsnp', username 'admin'; error -2005.

I'm absolutely 100% sure I type in the correct login name and password. (login: admin, password: secret).
I've also used the server 'arcserve' and 'casdb', but no luck.

I've two other computers with Brightstor v11.1 installed and when I try this same command on those computers, it works!

Can it be that my database is really corrupt?? I can't even use Brightstor manager normally, nor the Admin program.

Any help is appreciated.

 

[speshalyst]

Hi,

YOu were using the wrong server name..

rdsnp is invalid.. use casdb..

dbcheck -a -L casdb;admin;secret [db name]


Cheers


So it Shall be Written!
So it Shall be Done!!

 

[segil]

Hi speshalyst,

thanks for your reply. I also tried the servername casdb, but it won't work either. Nor arcserve, nor rdsnp.

Last week the problems started when I rebooted the server while it was doing a backup. (the backup proces was very slow and wouldn't respond)
Since then I got error messages about the database and now I can't use dbcheck. The database engine is started though.

I'm afraid I need to remove and reinstall arcserve, without the databases, because they're corrupt.

 

[segil]

I've tried it once again and this is the result:

D:\CA\BrightStor ARCserve Backup>dbcheck -a -L casdb;admin;secret Astpsdat

DBCHECK Utility - Velocis 2.1.410 [12-Mar-2001]
Copyright (c) 1992-2001 Centura Software Corporation. All Rights Reserved.

Failed to log into server 'casdb', username 'admin': error -2005.

D:\CA\BrightStor ARCserve Backup>

It works on the two other computers with Brightstor v11.1

 

[speshalyst]

Hi,

Error 2005 ususally means wrong server name..

In the BAB folder look for a tool called admin.exe

Open this tool.
On the Server menu choose login.
On the dialog window that opens .. type in CASDB on server(it would have rdsnp by default)

admin
secret

try to login.. if this fails..

check for RNNAME\LNNAME\ -
did you change the name of the server recently ?
Check if rds.exe is running..



So it Shall be Written!
So it Shall be Done!!

 

[segil]

Okay, with admin.exe I can't logon... it says:
"unable to get server connection"

RNNAME/LNNAME variables aren't listed in Windows.

rds.exe isn't running. Yesterday this proces consumed 100% cpu time continually. I killed it and renamed all files in the temp folder and one rAAAAAANA.chg file in the database folder, because this file caused rds.exe to go to 100% cpu time.... but now it's not running anymore... when I run it manually I get the following output:

D:\CA\BrightStor ARCserve Backup>rds

Velocis Server - Version 2.1.410 [12-Mar-2001]
Copyright (c) 1992-2001 Centura Software Corporation. All Rights Reserved.

[25-Oct-2005 15:05:36] CATPATH variable D:\CA\BrightStor ARCserve Backu
p\DATABASE
[25-Oct-2005 15:05:37] RECOVERY REQUIRED
[25-Oct-2005 15:05:37] RECOVERY: change log file=D:\CA\BrightStor ARCserve Back
up\DATABASE\rAAAAANA.chg
[25-Oct-2005 15:05:37] reopenFile: errno=2: failed to open file 'D:\CA\BrightSt
or ARCserve Backup\DATABASE\rAAAAANA.chg'
[25-Oct-2005 15:05:37] ***********************************
[25-Oct-2005 15:05:37] * Please report this condition to *
[25-Oct-2005 15:05:37] * your software supplier. *
[25-Oct-2005 15:05:37] ***********************************
[25-Oct-2005 15:05:37] Exiting to operating system

D:\CA\BrightStor ARCserve Backup>

I'm afraid if I rename the rAAAAAANA.chg back again, it will cause rds.exe to go to 100% cpu-time once again.

 

[speshalyst]

Hi,

Guess your ARcserve DB is in bad shape..
I'd replace it with a blank copy of the DB.!!



So it Shall be Written!
So it Shall be Done!!

 

[segil]

Hello,

I also posted the output of rds.exe, once the AAAA file is back in business... guess my post didn't come through.. here's the output again.

I'm thinking about reinstalling Arcserve v11.1 completely. As far as I know the databases are backupped to tape also, every night. So does this mean I can restore the database from tape and have everything the way it was before Arcserve crashed? Or maybe I can also use this installing and restore the database from tape and place it in the database folder, overwriting the corrupt database... will that solve all my problems???? Just remove the database folder and restore it from tape? Or is it not as simple as that?




Output:

Velocis Server - Version 2.1.410 [12-Mar-2001]
Copyright (c) 1992-2001 Centura Software Corporation. All Rights Reserved.

[25-Oct-2005 15:10:05] CATPATH variable D:\CA\BrightStor ARCserve Backu
p\DATABASE
[25-Oct-2005 15:10:05] RECOVERY REQUIRED
[25-Oct-2005 15:10:05] RECOVERY: change log file=D:\CA\BrightStor ARCserve Back
up\DATABASE\rAAAAANA.chg
[25-Oct-2005 15:10:05] RECOVERY: pass 1, discover last checkpoint address
[25-Oct-2005 15:10:06] RECOVERY: checkpoint address=9777492, EOF=10064186
[25-Oct-2005 15:10:06] RECOVERY: pass 2, build REDO and UNDO list
[25-Oct-2005 15:10:06] RECOVERY: pass 3, redo 34 transactions
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1106
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1107
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1108
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1109
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1110
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1111
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1112
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1113
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1114
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1115
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1116
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1117
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1118
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1119
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1120
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1121
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1122
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1123
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1124
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1125
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1126
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1127
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1128
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1129
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1130
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1131
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1132
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1133
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1134
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1135
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1136
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1137
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1138
[25-Oct-2005 15:10:06] RECOVERY: REDO transaction 1139
[25-Oct-2005 15:10:06] RECOVERY: opened database astpsdat
[25-Oct-2005 15:10:06] ***Fatal runtime library error***
[25-Oct-2005 15:10:06] Specific problem: Unable to read database name from
catalog
[25-Oct-2005 15:10:06] Task #: 1, User name: servertask
[25-Oct-2005 15:10:06] Current database handle: 65537, name: catalog
[25-Oct-2005 15:10:06] Function Entry: RUNTIME\DBD.C:406
[25-Oct-2005 15:10:06] Error Location: RUNTIME\RECOVER.C:666
[25-Oct-2005 15:10:06] ***********************************
[25-Oct-2005 15:10:06] * Please report this condition to *
[25-Oct-2005 15:10:06] * your software supplier. *
[25-Oct-2005 15:10:06] ***********************************
[25-Oct-2005 15:10:06] Exiting to operating system

 

[speshalyst]

Hi,

Reinstalling the application is good. But if you had backed up a corrupt DB.. you'd be putting back a corrupt DB..

i'd prefer to start clean..

So it Shall be Written!
So it Shall be Done!!

 

[segil]

I guess the db became corrupted last week, when I rebooted the server when I was doing some merging and other database activities... I could recover the db and run dbcheck to see if everything works okay... or do you think that's a bad idea?

If I decide to start with a blank db, how do I do that? Just erase the entire database folder contents and let arcserve figure create a new one?

 

[speshalyst]

Hi,

If you reinstall ARCserve it would create a new DB..

Else, install ARCserve on a different machine and copy the DB folder over to this server..

However, if you have any GFS rotation jobs, you'd have to recreate them as the media pool info would be lost..


So it Shall be Written!
So it Shall be Done!!

 

[segil]

okay... thanks for all your help.
Tonight I'm going to try to recover the db from tape and see if that works... otherwise I'll completely remove/reinstall the software... I'm also going to connect the tape streamer to a second scsi card, because it's giving a lot of trouble when connected to the first card, which also controls the harddisks.
Hopefully it will work at the end of the evening!

 

[segil]

Well.... it worked!!! I've removed Arcserve completely, installed a second scsi card and attached the tape device to this card. Re-installed arcserve with all v11.1 SP1 and it works like a charm! Speed has also increased, max speed is now 1.2 GB/min... not bad!
The 120 GB is backup-d in 4 hours now.

(recovering the database from tape didn't work... the manager constantly crashed)

Now I also know when I did wrong last time... a normal empty database contains a lot of small files (with an old time stamp)... why's that? The day before yesterday I removed these old files, because I thought it was old junk... guess I was wrong! *woeps*.... what do these aslogerr.xxx and asmsg.xxx and astpsdat.xxx do?? They're all empty and have a very old time stamp...Well, I now know not to touch these files anymore!