Thanks Robert!

That could very well be what is happening here.  Similarly, when I notice
Samba is eating a lot of CPU, I kill it and restream it.

I've downloaded the 2.2.4 version of Samba (thanks for the port Lars!) and
will probably be installing it on or about July 8th (after the holidays!).

Who else is running 2.2.4?

John


At 2002-06-28 07:11 AM, Robert Mills wrote:
>From: Robert Mills <[log in to unmask]>
>To: 'John Korb' <[log in to unmask]>, [log in to unmask]
>Subject: RE: Samba version 2.0.10 feedback
>Date: Fri, 28 Jun 2002 12:11:42 +0100
>MIME-Version: 1.0
>X-Mailer: Internet Mail Service (5.5.2653.19)
>Content-Type: multipart/alternative;
>         boundary="----_=_NextPart_001_01C21E94.9558C0D0"
>X-RCPT-TO: <[log in to unmask]>
>X-UIDL: 324929009
>Status: U
>
>
>---------->> John Korb said:
>Every once in a while (it has happened five times in four months) a Samba
>process apparently decides to loop and eats as much of the CPU as it
>can.  One rather idle day (our production hours are from mid afternoon to
>oh dark thirty) I decided to let it run for a few hours to see if it would
>eventually terminate or return to good behavior.  No such luck.
>
>I don't know what causes the looping, and it is infrequent, so I can
>easily live with it.  I'm wondering if anyone else using Samba 2.0.10 has
>experienced this.
>
>---------->>
>
>I've had this happen with version 2.0.7 on many occasions on our 979/400
>where CPU can go up to 96%.
>
>We first thought that it was because we were running Samba under control
>of INETD. So we pulled it out of there and ran the JSMB207 and JNMB207
>jobs, from SAMBA.SYS, which made no difference.
>
>Further investigation revealed that it only occurred when a PC application
>had problems with some data contained in the file it was importing. The
>application threw up a dialog box indicating what it didn't like and then
>terminated the import process (think that it's not closing the file before
>killing the import process). Samba, not knowing that it has finished with
>the file, just sits there waiting for the next read.
>
>Extrapolating, from information obtained in other postings on this list, I
>think that MPE assumes that Samba is stalled due to lack of resources and
>thus keeps increasing it's CPU priority to try and clear it thus
>compounding the problem.
>
>Our current work-around is to keep an eye on the CPU level for Samba and
>when it gets to high we kill and restream the JSMB207 job.
>
>regards,
>
>Robert W.Mills
>Systems Development Manager
>Windsong Services

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *