Smalltalk › Squeak › Squeak - Dev

DNU on processing update-ar101.mcm

_‹ Previous Topic Next Topic _›

Classic

List

Threaded

9 messages Options

Randal L. Schwartz

DNU on processing update-ar101.mcm

MCDiffyVersion>>baseSnapshot

is trying to send "nil snapshot".

Oops?

--
Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095
<[hidden email]> <URL:http://www.stonehenge.com/merlyn/>
Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc.
See http://methodsandmessages.vox.com/ for Smalltalk and Seaside discussion

Andreas.Raab

Re: DNU on processing update-ar101.mcm

Randal L. Schwartz wrote:
> MCDiffyVersion>>baseSnapshot
>
> is trying to send "nil snapshot".
>
> Oops?

I thought that Monticello-nice.342.mcz had taken care of that:

==================== Summary ====================

Name: Monticello-nice.342
Author: nice
Time: 27 December 2009, 9:03:54 am
UUID: ead29222-29d3-421a-8d89-08d667014c1c
Ancestors: Monticello-bf.341

change #findSnapshotWithVersionInfo: to silently ignore a missing
ancestor rather than sending #snapshot to nil.
This happens for example when merging Pharo/CollectionTests.
Don't know if this is the good strategy though (maybe a notification
could be signalled)

+ Cosmetic: move or remove a few temps inside closures

=================================================

Generally, when something like this happens, it's a good idea to open an
MC browser do a "flush cached versions" and try again. Most times it'll
just work.

Cheers,
- Andreas

Randal L. Schwartz

Re: Re: DNU on processing update-ar101.mcm

>>>>> "Andreas" == Andreas Raab <[hidden email]> writes:

Andreas> Randal L. Schwartz wrote:
MCDiffyVersion> baseSnapshot
>> is trying to send "nil snapshot".
>> Oops?

Andreas> I thought that Monticello-nice.342.mcz had taken care of that:

Well, I'm reporting what happens, in hopes that it rings a bell.

Andreas> Generally, when something like this happens, it's a good idea to open an MC
Andreas> browser do a "flush cached versions" and try again. Most times it'll just work.

So, some of the files on my hard drive should be removed before updating?
If this is true, why isn't this done automatically?

As in "load code updates" should automatically do "flush cached versions" as a
matter of policy, if it'll help naive people move forward.

--
Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095
<[hidden email]> <URL:http://www.stonehenge.com/merlyn/>
Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc.
See http://methodsandmessages.vox.com/ for Smalltalk and Seaside discussion

Andreas.Raab

Re: DNU on processing update-ar101.mcm

Randal L. Schwartz wrote:
> Andreas> Generally, when something like this happens, it's a good idea to open an MC
> Andreas> browser do a "flush cached versions" and try again. Most times it'll just work.
>
> So, some of the files on my hard drive should be removed before updating?
> If this is true, why isn't this done automatically?
>
> As in "load code updates" should automatically do "flush cached versions" as a
> matter of policy, if it'll help naive people move forward.

Good point and trivial to do. Just change, i.e.,

Utilities class>>updateFromServer
"Update the image by loading all pending updates from the server."
| config |
"Flush all caches for good measure before updating.
If a previous download failed this is often helpful."
MCFileBasedRepository flushAllCaches.
config := MCMcmUpdater updateFromRepositories: #(
'http://source.squeak.org/trunk'
).
self setSystemVersionFromConfig: config.
self inform: 'Update completed.
Current update number: ', SystemVersion current highestUpdate.

Unfortunately, it appears source.squeak.org is down right now. I'm
getting timeouts and proxy errors.

Cheers,
- Andreas

Ken Causey-3

source.squeak.org restarted (was Re: Re: DNU on processing update-ar101.mcm)

On Sat, 2010-01-02 at 18:52 +0100, Andreas Raab wrote:
> Unfortunately, it appears source.squeak.org is down right now. I'm
> getting timeouts and proxy errors.
>
> Cheers,
> - Andreas

Yep, it was locked hard. I tried a few things but ultimately resorted
to killing it and restarting. It had last saved the image around 13:15
UTC on the 2nd, and at a glance it doesn't appear that anything was
lost. But if you have done anything in the last 24 hours please take a
look.

Ken

signature.asc (197 bytes) Download Attachment

Bert Freudenberg

Re: source.squeak.org restarted (was Re: Re: DNU on processing update-ar101.mcm)

On 02.01.2010, at 23:27, Ken Causey wrote:

>
> On Sat, 2010-01-02 at 18:52 +0100, Andreas Raab wrote:
>> Unfortunately, it appears source.squeak.org is down right now. I'm
>> getting timeouts and proxy errors.
>>
>> Cheers,
>> - Andreas
>
> Yep, it was locked hard. I tried a few things but ultimately resorted
> to killing it and restarting. It had last saved the image around 13:15
> UTC on the 2nd, and at a glance it doesn't appear that anything was
> lost. But if you have done anything in the last 24 hours please take a
> look.
>
> Ken

Locked up again, last saved 11:15 UTC on the 3rd.

It seems to be running though. I had a look in gdb and it was interpreting and gc'ing but printAllStacks() didn't print anything. Strace agrees it's doing something:

squeaksource@box2:~$ strace -p 5314
read(9, "", 2000) = 0
gettimeofday({1262521841, 160596}, NULL) = 0
gettimeofday({1262521841, 160658}, NULL) = 0
gettimeofday({1262521841, 160706}, NULL) = 0
read(9, "", 2000) = 0
gettimeofday({1262521841, 160813}, NULL) = 0
gettimeofday({1262521841, 160873}, NULL) = 0
gettimeofday({1262521841, 160922}, NULL) = 0
squeaksource@box2:~$ ll /proc/5314/fd/9
lrwx------ 1 squeaksource squeaksource 64 Jan 3 12:31 /proc/5314/fd/9 -> socket:[128749641]

Restarted at 12:39.

- Bert -

Ken Causey-3

Re: source.squeak.org restarted (was Re: Re: DNU on processing update-ar101.mcm)

On Sun, 2010-01-03 at 13:42 +0100, Bert Freudenberg wrote:

> On 02.01.2010, at 23:27, Ken Causey wrote:
> >
> > On Sat, 2010-01-02 at 18:52 +0100, Andreas Raab wrote:
> >> Unfortunately, it appears source.squeak.org is down right now. I'm
> >> getting timeouts and proxy errors.
> >>
> >> Cheers,
> >> - Andreas
> >
> > Yep, it was locked hard. I tried a few things but ultimately resorted
> > to killing it and restarting. It had last saved the image around 13:15
> > UTC on the 2nd, and at a glance it doesn't appear that anything was
> > lost. But if you have done anything in the last 24 hours please take a
> > look.
> >
> > Ken
>
> Locked up again, last saved 11:15 UTC on the 3rd.
>
> It seems to be running though. I had a look in gdb and it was interpreting and gc'ing but printAllStacks() didn't print anything. Strace agrees it's doing something:
>
> squeaksource@box2:~$ strace -p 5314
> read(9, "", 2000) = 0
> gettimeofday({1262521841, 160596}, NULL) = 0
> gettimeofday({1262521841, 160658}, NULL) = 0
> gettimeofday({1262521841, 160706}, NULL) = 0
> read(9, "", 2000) = 0
> gettimeofday({1262521841, 160813}, NULL) = 0
> gettimeofday({1262521841, 160873}, NULL) = 0
> gettimeofday({1262521841, 160922}, NULL) = 0
> squeaksource@box2:~$ ll /proc/5314/fd/9
> lrwx------ 1 squeaksource squeaksource 64 Jan 3 12:31 /proc/5314/fd/9 -> socket:[128749641]
>
> Restarted at 12:39.
>
>
> - Bert -

And the saga continues. I found it down today (1/3/10) and using your
comments above tried to examine the situation but found no more
information than you did. For what it's worth there was a progress
display up saying 'Scanning ReadStream...'.

So I killed it and started it back up, but the problem was still there,
apparently the image had managed to save (at 15:15) with the problem.
My only choice seemed to be to rename the current copy and restore the
most recent from backup (saved at 03:15). This older copy is now
running.

It is almost certainly the case that something was lost this time.

Ken

signature.asc (197 bytes) Download Attachment

Andreas.Raab

Re: source.squeak.org restarted (was Re: Re: DNU on processing update-ar101.mcm)

Ken Causey wrote:

> And the saga continues. I found it down today (1/3/10) and using your
> comments above tried to examine the situation but found no more
> information than you did. For what it's worth there was a progress
> display up saying 'Scanning ReadStream...'.
>
> So I killed it and started it back up, but the problem was still there,
> apparently the image had managed to save (at 15:15) with the problem.
> My only choice seemed to be to rename the current copy and restore the
> most recent from backup (saved at 03:15). This older copy is now
> running.

Any chance you can make a copy of the image/data in the failed state for
a post mortem? This is getting annoying so knowing what fails would be
useful.

Cheers,
- Andreas

Levente Uzonyi-2

Re: source.squeak.org restarted (was Re: Re: DNU on processing update-ar101.mcm)

In reply to this post by Ken Causey-3

On Mon, 4 Jan 2010, Ken Causey wrote:

What about building a new image, based on a trunk image for example?
Could it update itself? :)

Levente

>
> Ken
>