Translating with DrWatson… this can take a few seconds the first time.
This is a genuine, complex translation. DrWatson protects commands, error codes, and log output while naturally translating the surrounding text. It’s translated once and saved.
Carlson@WHSmith — — source: Informix-list mailing list archive (1991-1998)
Bill Weaver wrote:
>
> Today for some reason we've added 5 shared memory segments dynamically. It almost never happens
> that we even have 1 added since I've worked to size our initial segment to avoid this. How can
> I tell what is causing us to add so many all of a sudden?
>
Index creations are the 'big' one on our systems. I believe that
sorting also occurs in virtual memory.
According to the manual, big buffers, session and thread data are also
stored in virtual memory.
> Also our ready queue seems to be long today (we almost never have any processes in it when we do
> onstat -g rea) even though our processor utilization is only around 60%. The cpuvps are> affinitied to physical processors (there are 8 cpus in the box and we run 6 cpuvps). Could this
> be related to the problem above? What else could cause it? Would adding another cpuvp help
> even though the physical cpus aren't at capacity?
Are you running on HP? There exists a known problem with how HP
utilizes more than three <?> active memory segments, so I watch my
segments using onstat -g seg. If we continue to reallocate additional
segments after a reboot, then I will increase SHMVIRTSIZE accordingly
John Carlson
Informix DBA
WH Smith, Inc.
We use strictly necessary cookies to make this site work. With your
consent we’d also use optional cookies for analytics and marketing. You can accept all,
reject all, or choose. Read our Cookie Policy.