buzzardj wrote:

>
>
> Hi,
>
> I am still battling with this problem.
> I have noticed during a freeze there is always at least 1 firebird
> process with the flag 0x70 in state hung.
>
> OWNER BLOCK 27936
> Owner id: 27997, type: 3, flags: 0x30, pending: 0,
> semid: 15 (available)
> Process id: 27997, UID: 0x1F5 Alive
> Flags: 0x70 hung wake
> Requests (284): forward: 752284, backward: 151524
> Blocks (1): forward: 242320, backward: 242320
>
> I have found that after killing that process, firebird usually
> recovers and will once again accept new connections.
>
> In a freeze today there were 6 hung processes with the flag 0x70 and
> approx. 50 others in various hung states.
> Flags: 0x40 hung
> Flags: 0x44 hung
> Flags: 0x60 hung wake
>
> After killing one 0x70 process all the other 0x70 and hung processes
> disappeared and firebird came back to life.
>
> Does anyone know what these flags are, 0x70 specifically?
> Any ideas on how I can track the source of these freezing issues down?
>
> Many thanks,
> Richard

Just guessing here..but could it be related to linux threading problems
? I would check this first.


Regards
Boguslaw