<div dir="ltr">Hi David,<div class="gmail_extra"><br><div class="gmail_quote">On Thu, Nov 14, 2013 at 6:00 PM, David T. Lewis <span dir="ltr"><<a href="mailto:lewis@mail.msen.com" target="_blank">lewis@mail.msen.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Attached is a screen shot of the process browser in the <a href="http://squeaksource.com" target="_blank">squeaksource.com</a><br>
image, showing the excess SSSession processes. They are deadlocked on<br>
accessing DateAndTime now, which contains a critical section using the<br>
LastTickSemaphore in class DateAndTime.<br>
<br>
In the <a href="http://squeaksource.com" target="_blank">squeaksource.com</a> image, LastTickSemaphore has 0 excess signals,<br>
whereas in other images I look at, it has 1 excess signal. This looks<br>
to me like a mutex that has gotten confused.<br>
<br>
I sent a signal to LastTickSemaphore in class DateAndTime, and now it<br>
looks like a mutex again. Let's see if that clears the problem.<br>
<br>
This certainly has a bad smell about it :-( But I note also that<br>
we are running our SqueakSource services on older images, and a number<br>
of changes have been made to DateAndTime since then.<br>
<br>
Nicolas, I will send private email to give you the VNC password for access<br>
to the <a href="http://squeaksource.com" target="_blank">squeaksource.com</a> image in case you need it (I am going to get some<br>
sleep soon).<br></blockquote><div><br></div><div>All this LastTickSemaphore stuff is complete nonsense, wasting on average 1/2 a second on startup spinning until the clock rolls over. If we move to the 64-bit microsecond timebase which is provided by the Cog time primitives we don't need to sync the second and the millisecond clocks because they are replaced by a single microsecond clock. If the current Interpreter VMs do support the 64-bit microsecond primitives I suggest we move ASAP. QWe've already done this in our images at Cadence and been running happily with it for several months. Would this help?</div>
<div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
Dave<br>
<br>
<br>
On Thu, Nov 14, 2013 at 08:26:45PM -0500, David T. Lewis wrote:<br>
> The image is now showing 10 session handler processes in the process<br>
> browser. Presumably these are related to the failed upload requests.<br>
><br>
> I do not understand the cause of this problem, and it may be that I<br>
> should revert the changes that I did earlier today (in which I put<br>
> squeaksource under the control of the supervise(8) for starting the<br>
> image).<br>
><br>
> But I suspect that the problem lies elsewhere, so for I will make<br>
> a copy of the broken image for debugging, then terminate the excess<br>
> processes. This should clear the problem temporarily. I will follow<br>
> up with another email within about 30 minutes.<br>
><br>
> Dave<br>
><br>
><br>
><br>
> On Fri, Nov 15, 2013 at 01:28:18AM +0100, Nicolas Cellier wrote:<br>
> > Thanks David, it went back to normal speed for a moment, but is now<br>
> > rejecting my upload requests again (most will timeout, some do work<br>
> > intermittently)...<br>
> ><br>
> ><br>
> > 2013/11/14 David T. Lewis <<a href="mailto:lewis@mail.msen.com">lewis@mail.msen.com</a>><br>
> ><br>
> > > On Thu, Nov 14, 2013 at 10:55:55PM +0100, Nicolas Cellier wrote:<br>
> > > > Yes, let's remove the alarms.<br>
> > > > But it has to be functional.<br>
> > > > Currently, I can connect on the web interface and I can download, but all<br>
> > > > my upload are failing with timeout... Any idea?<br>
> > > ><br>
> > ><br>
> > > I made some changes to the launch script for <a href="http://squeaksource.com" target="_blank">squeaksource.com</a> earlier<br>
> > > today:<br>
> > ><br>
> > ><br>
> > > <a href="http://lists.squeakfoundation.org/pipermail/box-admins/2013-November/001598.html" target="_blank">http://lists.squeakfoundation.org/pipermail/box-admins/2013-November/001598.html</a><br>
> > ><br>
> > > It is possible that this may be related to the problem you are seeing (I am<br>
> > > not sure at this point).<br>
> > ><br>
> > > I tried loading some packages from squeaksource a few minutes ago, and it<br>
> > > was slow but functional. However, checking the image I see 15 active<br>
> > > SSession handlers in a ProcessBrowser. This is not right, and it appears<br>
> > > to be a recurrence of a problem that we have seen previously on an<br>
> > > intermittent<br>
> > > basis, both on <a href="http://squeaksource.com" target="_blank">squeaksource.com</a> and (probably) on <a href="http://source.squeak.org" target="_blank">source.squeak.org</a>.<br>
> > ><br>
> > > I will terminate the runaway session handler processes, which I hope will<br>
> > > clear up the immediate problem.<br>
> > ><br>
> > > More to follow I'm sure ...<br>
> > ><br>
> > > Dave<br>
> > ><br>
> > > ><br>
> > > > 2013/11/14 Bert Freudenberg <<a href="mailto:bert@freudenbergs.de">bert@freudenbergs.de</a>><br>
> > > ><br>
> > > > ><br>
> > > > > On 14.11.2013, at 13:04, Chris Muller <<a href="mailto:asqueaker@gmail.com">asqueaker@gmail.com</a>> wrote:<br>
> > > > ><br>
> > > > > >> The welcome (sic) message on the SqueakSource home page is overly<br>
> > > > > alarming,<br>
> > > > > >> and IMHO should be changed to something that encourages new<br>
> > > projects to<br>
> > > > > >> be created elsewhere, but that does not cause alarm for existing<br>
> > > usurs.<br>
> > > > > >> But that is a policy decision, and I will defer to the Squeak board<br>
> > > and<br>
> > > > > >> the Squeak community on this.<br>
> > > > > ><br>
> > > > > > +1<br>
> > > > > ><br>
> > > > > > I think we should we should delete the "ATTENTION!" line but leave<br>
> > > the<br>
> > > > > > note about creation of projects being disabled.<br>
> > > > ><br>
> > > > > Yep.<br>
> > > > ><br>
> > > > > - Bert -<br>
> > > > ><br>
> > > > ><br>
> > > > ><br>
> > > > ><br>
> > ><br>
> > > ><br>
> > ><br>
> > ><br>
> > ><br>
><br>
> ><br>
<br><br>
<br></blockquote></div><br><br clear="all"><div><br></div>-- <br>best,<div>Eliot</div>
</div></div>