Re: [Info-ingres] E_SC0107_BAD_SIZE_EXPAND ??
- From: Karl & Betty Schendel <schendel@xxxxxxxxxxxxxx>
- Date: Tue, 12 Sep 2006 15:15:57 -0400
At 2:59 PM -0400 9/12/06, Dennis Roesler wrote:
On HP-UX 11x / Ingres II 3.0.2 (hp2.us5/109)
About once a week we get an error
E_SC0107_BAD_SIZE_EXPAND Error expanding virtual size of server.
brk() failed with operating system error 12 (Not enough space).
Bouncing ingres resolves this, but what is it trying to expand and where, so I can see where more space needs to be added?
If you don't have II_DBMS_LOG set, try setting it to an absolute
pathname (e.g. $II_SYSTEM/ingres/files/iidbms.log or iidbms_%p.log
to append the dbms process id). Use ingsetenv.
Although it sometimes takes a little deciphering, the dbms log
should tell you what facility bumped its head on the ceiling.
An SC0107 unaccompanied by another XXX out of memory message can
usually be blamed on DMF. (Are you using partitioned tables with
thousands of partitions simultaneously active across all sessions?
That can be a memory eater in 3.0.2.)
If it's a complete mystery, it might possibly be thread stacks.
I haven't worked with the HP/UX port so I'm not sure whether
it's obeying the stack_size parameter. It is supposed to,
but early in r3 the linux port wasn't, so who knows...
Karl
.
- Follow-Ups:
- Re: [Info-ingres] E_SC0107_BAD_SIZE_EXPAND ??
- From: Dennis Roesler
- Re: [Info-ingres] E_SC0107_BAD_SIZE_EXPAND ??
- References:
- E_SC0107_BAD_SIZE_EXPAND ??
- From: Dennis Roesler
- E_SC0107_BAD_SIZE_EXPAND ??
- Prev by Date: Re: Deciphering errors
- Next by Date: Re: [Info-ingres] E_SC0107_BAD_SIZE_EXPAND ??
- Previous by thread: E_SC0107_BAD_SIZE_EXPAND ??
- Next by thread: Re: [Info-ingres] E_SC0107_BAD_SIZE_EXPAND ??
- Index(es):