[p4] Anyone experienced these perforce problems? - Rajesh Vaidheeswarran - com.perforce.perforce-user

1 message in com.perforce.perforce-user[p4] Anyone experienced these perforc...

From	Sent On	Attachments
Rajesh Vaidheeswarran	12 Jul 2001 12:53

Subject:	[p4] Anyone experienced these perforce problems?
From:	Rajesh Vaidheeswarran (rv...@nauticusnet.com)
Date:	07/12/2001 12:53:22 PM
List:	com.perforce.perforce-user

To close on this one,

We moved the perforce depot back from the Netapp to a local disk on the sun, and the problems went away.

This is really sad because a netapp would've helped the backup situation tremendously. But, this really brought out one thing for us... sun and netapp don't play well on NFSv3 for critical apps like perforce.

It might be a combination of hardware that we have that is the culprit and not the two boxes themselves... but it looks better the way we have it now.

Thanks for all the suggestions.

-----Original Message----- From: Chuck Karish [mailto:karish at well.com] Sent: Wednesday, June 27, 2001 3:16 PM To: Rajesh Vaidheeswarran; 'perforce-user at perforce.com' Subject: RE: [p4] Anyone experienced these perforce problems?

At 02:02 PM 6/27/2001 -0400, Rajesh Vaidheeswarran wrote:

We have been encountering *very long* delays from the client side for many p4 commands .. a p4 edit which typically returns in under a second returns at times after 10 minutes! Most often it times out..

What times out? My experience is that neither the Perforce server nor the client time out on their own. Are attempts to open connections timing out after four minutes?

The client side errors out with something like this. But its not always the same error. I'm collecting the various types of errors that we are seeing.

Perforce client error: Connection with partner closed unexpectedly. TCP receive failed. read: socket: WSAECONNRESET

That's the message you'd see if the user were to terminate the connection from the client side. Are you sure that they're not doing that?

The messages you showed us all look like reports of client/server problems, not server/fileserver problems. Many of them are the usual messages I expect to see when a user interrupts a long-running client process or when a bad client spec is used (non-existent local Root).

How do you say so? It looks like they are client-server problems, but I am not sure if they are (or are not) server/fileserver problems too.

The messages about access denied refer to files in the client workspace, not in the depot.