Subject | Re: FireBird 1.0 server crash with stack overflow |
---|---|
Author | Aage Johansen |
Post date | 2002-11-18T20:43:43Z |
Riaan Nagel wrote:
you reduced the number of page buffers to something below 10 000 (if,
however, you find 40000 better just ignore this). Maybe a larger page size
would help also (e.g. 8KB).
Your setup looks fine, but if there are several databases opened
simultaneously your RAM (256MB) may be on the small side.
The sudden increase in memory may indicate a 'stuck' transaction. Take a
look at the header page, and report back with the numbers for
Oldest transaction
Oldest active
Oldest snapshot
Next transaction
Regards,
Aage J.
> One of our dedicated Win2K FireBird 1.0 servers crashed a few days=20This won't help with your problem, but you might have better performance if
> ago, without any entry in the interbase.log and with only the=20
> following message in the System Event Viewer:
> ... ...
> Database is set up to not do automatic sweeping, has forced writes=20
> turned on, 4KB page size, 40 000 page buffers, is around 1GB in size,=20
> in a single file, with no shadows and no replication and is backed up=20
> every evening. We are not currently doing a manual sweep.
> The server configurations are as follows:=20
> Processor: 1.2GHz
> Physical Memory: 256MB
> Virtual Memory: 384MB
> Nic: Intel Pro100 S (x2)
> Windows 2000 SP3 Running IIS5
> Software mirroring is done to one other hard drive.
you reduced the number of page buffers to something below 10 000 (if,
however, you find 40000 better just ignore this). Maybe a larger page size
would help also (e.g. 8KB).
Your setup looks fine, but if there are several databases opened
simultaneously your RAM (256MB) may be on the small side.
The sudden increase in memory may indicate a 'stuck' transaction. Take a
look at the header page, and report back with the numbers for
Oldest transaction
Oldest active
Oldest snapshot
Next transaction
Regards,
Aage J.