Is anyone else looking into a problem that appears to have been introduced by the server change on Feb 1st changing thread and process ids to small integers?
As a result ACCPAC for Windows' main desktop process shuts down after about a minute, but leaves various other processes running.
I'll have a look at it but I don't know how successful I'll be.
"Bill" == Bill Medland billmedland@mercuryspeed.com writes:
Bill> Is anyone else looking into a problem that appears to have been Bill> introduced by the server change on Feb 1st changing thread and Bill> process ids to small integers?
Bill> As a result ACCPAC for Windows' main desktop process shuts down Bill> after about a minute, but leaves various other processes running.
Bill> I'll have a look at it but I don't know how successful I'll be.
Webpack doesn't work as before since the last CVS update. It uses a lot of processes and perhaps threads. Maybe this problem is correlated.
Bye
On February 13, 2003 11:42 am, Uwe Bonnes wrote:
"Bill" == Bill Medland billmedland@mercuryspeed.com writes:
Bill> Is anyone else looking into a problem that appears to have been Bill> introduced by the server change on Feb 1st changing thread and Bill> process ids to small integers? Bill> As a result ACCPAC for Windows' main desktop process shuts down Bill> after about a minute, but leaves various other processes running. Bill> I'll have a look at it but I don't know how successful I'll be.Webpack doesn't work as before since the last CVS update. It uses a lot of processes and perhaps threads. Maybe this problem is correlated.
Bye
Could be.
Since the last CVS update? Do you mean the last snapshot on 20030115? If not then it probably isn't; this problem was introduced two weeks ago. (Shows how much attention I am paying)
It looks like something goes wrong in the next_process handler. I'll let you know.
Ah ha, so that's the problem is it? I'm seeing a similar issue with IE6 after loading the adobe svg viewer plugin. For no apparent reason, Wine bombs after about 1-2 minutes. It doesn't crash or anything, it just stops, and the logs don't show anything out of the ordinary. The windows stay on the screen, but the processes appear hung.
How did you determine it was changing the ids? Just a regression search?
On Thu, 2003-02-13 at 18:52, Bill Medland wrote:
Is anyone else looking into a problem that appears to have been introduced by the server change on Feb 1st changing thread and process ids to small integers?
As a result ACCPAC for Windows' main desktop process shuts down after about a minute, but leaves various other processes running.
I'll have a look at it but I don't know how successful I'll be.
On February 13, 2003 11:56 am, Mike Hearn wrote:
Ah ha, so that's the problem is it? I'm seeing a similar issue with IE6 after loading the adobe svg viewer plugin. For no apparent reason, Wine bombs after about 1-2 minutes. It doesn't crash or anything, it just stops, and the logs don't show anything out of the ordinary. The windows stay on the screen, but the processes appear hung.
How did you determine it was changing the ids? Just a regression search?
Yep. It just takes processor time really, so that is the easiest way. Looks like it is the next_process handler.
On Thu, 2003-02-13 at 18:52, Bill Medland wrote:
Is anyone else looking into a problem that appears to have been introduced by the server change on Feb 1st changing thread and process ids to small integers?
As a result ACCPAC for Windows' main desktop process shuts down after about a minute, but leaves various other processes running.
I'll have a look at it but I don't know how successful I'll be.
Bill Medland wrote:
Is anyone else looking into a problem that appears to have been introduced by the server change on Feb 1st changing thread and process ids to small integers?
does this help ? A+
On February 13, 2003 12:29 pm, Eric Pouech wrote:
Bill Medland wrote:
Is anyone else looking into a problem that appears to have been introduced by the server change on Feb 1st changing thread and process ids to small integers?
does this help ? A+
Yep.
(So should the fix be there or should get_process_id be able to handle a null pointer?)
Bill Medland wrote:
On February 13, 2003 12:29 pm, Eric Pouech wrote:
Bill Medland wrote:
Is anyone else looking into a problem that appears to have been introduced by the server change on Feb 1st changing thread and process ids to small integers?
does this help ? A+
Yep.
(So should the fix be there or should get_process_id be able to handle a null pointer?)
since the function is inline, it doesn't matter much I'd better leave it that way, so it makes it clear that we don't always keep track of the parent for a given process A+
-
Bill Medland -
Eric Pouech -
Mike Hearn -
Uwe Bonnes