Discussion:
6 tests to go...
(too old to reply)
Frank Shearar
2012-10-23 00:21:51 UTC
Permalink
We're down to 6 failing unit tests on the Linux platform, one of which
is an expected failure and another of which is a known network issue
for the Linux platform.

Would my kind Windows and OSX build volunteers please run builds
against the latest trunk image?
(http://173.246.101.237:8080/job/SqueakTrunk/lastCompletedBuild/artifact/)

Note the new URL: that's thanks to Chris Cunnington, who has kindly
moved Jenkins to its new home. That will no doubt shortly become the
new squeakci.org.

Windows users note please that you'll need to grant permission for the
build executables etc to act as servers and to make connections: you
can do this through the Windows Firewall configuration screens. (I'd
give more detailed instructions, but my Windows machines are far away
right now.)

(Dale, this shouldn't affect you much, in the sense that you won't
need any changes to builderCI or whatever once the DNS changes have
been made. So if they _do_ affect you, they shouldn't affect you for
very long.)

It's about time I started with the more admin-y side of the release,
making a Changelog and such. I'll make a start on this soon,
docs/progress to be on http://wiki.squeak.org/squeak/6188

frank
Tom Rushworth
2012-10-23 21:34:16 UTC
Permalink
Hi Frank,

I made a stab at getting the new image to run on my OS X server, but the
link below doesn't seem to do anything in a regular browser, or when I
paste it into the update-image.st script as the update source.

I'm probably completely misunderstanding what I should be doing...
Can you give me a bigger hint? :).

BTW, your message fell prey to my Postini spam filter, so it got delayed
a day or so, until I spotted it in the quarantine summary. I've put
your address and squeak-dev on my Postini approved list, so it shouldn't
happen again.
Post by Frank Shearar
We're down to 6 failing unit tests on the Linux platform, one of which
is an expected failure and another of which is a known network issue
for the Linux platform.
Would my kind Windows and OSX build volunteers please run builds
against the latest trunk image?
(http://173.246.101.237:8080/job/SqueakTrunk/lastCompletedBuild/artifact/)
Note the new URL: that's thanks to Chris Cunnington, who has kindly
moved Jenkins to its new home. That will no doubt shortly become the
new squeakci.org.
Windows users note please that you'll need to grant permission for the
build executables etc to act as servers and to make connections: you
can do this through the Windows Firewall configuration screens. (I'd
give more detailed instructions, but my Windows machines are far away
right now.)
(Dale, this shouldn't affect you much, in the sense that you won't
need any changes to builderCI or whatever once the DNS changes have
been made. So if they _do_ affect you, they shouldn't affect you for
very long.)
It's about time I started with the more admin-y side of the release,
making a Changelog and such. I'll make a start on this soon,
docs/progress to be on http://wiki.squeak.org/squeak/6188
frank
--
Tom Rushworth
Jeff Gonis
2012-10-23 21:39:23 UTC
Permalink
Hi Frank,

I return home tonight and I will have my Mac available to me at that
point. I will run the tests and post my results to the list. I could also
run them on my linux machine as well and post those results if no one has
beat me to it.

Jeff

On Tue, Oct 23, 2012 at 10:34 AM, Tom Rushworth
Post by Tom Rushworth
Hi Frank,
I made a stab at getting the new image to run on my OS X server, but the
link below doesn't seem to do anything in a regular browser, or when I
paste it into the update-image.st script as the update source.
I'm probably completely misunderstanding what I should be doing...
Can you give me a bigger hint? :).
BTW, your message fell prey to my Postini spam filter, so it got delayed
a day or so, until I spotted it in the quarantine summary. I've put
your address and squeak-dev on my Postini approved list, so it shouldn't
happen again.
Post by Frank Shearar
We're down to 6 failing unit tests on the Linux platform, one of which
is an expected failure and another of which is a known network issue
for the Linux platform.
Would my kind Windows and OSX build volunteers please run builds
against the latest trunk image?
(
http://173.246.101.237:8080/job/SqueakTrunk/lastCompletedBuild/artifact/)
Post by Frank Shearar
Note the new URL: that's thanks to Chris Cunnington, who has kindly
moved Jenkins to its new home. That will no doubt shortly become the
new squeakci.org.
Windows users note please that you'll need to grant permission for the
build executables etc to act as servers and to make connections: you
can do this through the Windows Firewall configuration screens. (I'd
give more detailed instructions, but my Windows machines are far away
right now.)
(Dale, this shouldn't affect you much, in the sense that you won't
need any changes to builderCI or whatever once the DNS changes have
been made. So if they _do_ affect you, they shouldn't affect you for
very long.)
It's about time I started with the more admin-y side of the release,
making a Changelog and such. I'll make a start on this soon,
docs/progress to be on http://wiki.squeak.org/squeak/6188
frank
--
Tom Rushworth
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20121023/bc363a61/attachment.htm
Frank Shearar
2012-10-23 22:15:32 UTC
Permalink
Post by Tom Rushworth
Hi Frank,
I made a stab at getting the new image to run on my OS X server, but the
link below doesn't seem to do anything in a regular browser, or when I
paste it into the update-image.st script as the update source.
I'm probably completely misunderstanding what I should be doing...
Can you give me a bigger hint? :).
Right. This is the second time we've seen this. There's _supposed_ to
be a zip there for you to download.

Chris Cunnington (I'm not saying that with my Dad Voice; there are
just lots of people sharing your name on the list), did you do
anything to the build server? It looks like all the build history for
SqueakTrunk just vanished.

frank
Post by Tom Rushworth
BTW, your message fell prey to my Postini spam filter, so it got delayed
a day or so, until I spotted it in the quarantine summary. I've put
your address and squeak-dev on my Postini approved list, so it shouldn't
happen again.
Post by Frank Shearar
We're down to 6 failing unit tests on the Linux platform, one of which
is an expected failure and another of which is a known network issue
for the Linux platform.
Would my kind Windows and OSX build volunteers please run builds
against the latest trunk image?
(http://173.246.101.237:8080/job/SqueakTrunk/lastCompletedBuild/artifact/)
Note the new URL: that's thanks to Chris Cunnington, who has kindly
moved Jenkins to its new home. That will no doubt shortly become the
new squeakci.org.
Windows users note please that you'll need to grant permission for the
build executables etc to act as servers and to make connections: you
can do this through the Windows Firewall configuration screens. (I'd
give more detailed instructions, but my Windows machines are far away
right now.)
(Dale, this shouldn't affect you much, in the sense that you won't
need any changes to builderCI or whatever once the DNS changes have
been made. So if they _do_ affect you, they shouldn't affect you for
very long.)
It's about time I started with the more admin-y side of the release,
making a Changelog and such. I'll make a start on this soon,
docs/progress to be on http://wiki.squeak.org/squeak/6188
frank
--
Tom Rushworth
Chris Cunnington
2012-10-23 22:19:31 UTC
Permalink
Post by Frank Shearar
Post by Tom Rushworth
Hi Frank,
I made a stab at getting the new image to run on my OS X server, but the
link below doesn't seem to do anything in a regular browser, or when I
paste it into the update-image.st script as the update source.
I'm probably completely misunderstanding what I should be doing...
Can you give me a bigger hint? :).
Right. This is the second time we've seen this. There's _supposed_ to
be a zip there for you to download.
Chris Cunnington (I'm not saying that with my Dad Voice; there are
just lots of people sharing your name on the list), did you do
anything to the build server? It looks like all the build history for
SqueakTrunk just vanished.
frank
The completed build you're saying missing is #17. Remember when I said
yesterday that I had a completed build called #16 and that it just...
disappeared. Seems to have happened again. I have no idea why. I logged
in and hit Build Now and that created #17. I just did that again. I
expect build #18 should be along presently. Why does this happen? I have
no idea.

Chris
Chris Cunnington
2012-10-23 22:21:31 UTC
Permalink
Post by Frank Shearar
Post by Tom Rushworth
Hi Frank,
I made a stab at getting the new image to run on my OS X server, but the
link below doesn't seem to do anything in a regular browser, or when I
paste it into the update-image.st script as the update source.
I'm probably completely misunderstanding what I should be doing...
Can you give me a bigger hint? :).
Right. This is the second time we've seen this. There's _supposed_ to
be a zip there for you to download.
Chris Cunnington (I'm not saying that with my Dad Voice; there are
just lots of people sharing your name on the list), did you do
anything to the build server? It looks like all the build history for
SqueakTrunk just vanished.
frank
Oh, and this is likely easier to use:

http://www.squeakci.org

Chris
Chris Cunnington
2012-10-23 23:00:41 UTC
Permalink
I think I can sketch out a few details.

1. The first thing is that this is not a medical emergency. I'm saying
that for my own peace of mind. System administration under duress is no
fun at all.

2. The builds are still there. They don't disappear. They just aren't
seen by the webpage. We are not alone in this. It has happened before.

http://jenkins.361315.n4.nabble.com/JIRA-JENKINS-11938-Jenkins-loses-builds-when-restarted-td4124991.html

3. The problem may be in the reading of an XML file. The solution I've
gleaned so far is that instead of creating another build, I could log in
and press:

"Reload Configuration from Disk"

Or I could just try restarting.

4. This didn't happen on the older server. It won't last forever on this
one.

5. God knows, somebody will come up with a better answer than I can. But
since you're all looking at me at this moment, I say this: "Don't panic.
We still know where our towel is."


Chris
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20121023/e50de333/attachment.htm
Igor Stasenko
2012-10-23 23:29:20 UTC
Permalink
Hi, Chris..

Just wanna add own 2 cents:

yes, Jenkins quality remains to be an issue.
We also experiencing problems with it time to time.

I wonder if there any better alternatives..
From other side, jenkins provides a rich set of features (many from plugins),
which would be hard to match if one would start writing it from scratch.

On 23 October 2012 20:00, Chris Cunnington
I think I can sketch out a few details.
1. The first thing is that this is not a medical emergency. I'm saying that
for my own peace of mind. System administration under duress is no fun at
all.
2. The builds are still there. They don't disappear. They just aren't seen
by the webpage. We are not alone in this. It has happened before.
http://jenkins.361315.n4.nabble.com/JIRA-JENKINS-11938-Jenkins-loses-builds-when-restarted-td4124991.html
3. The problem may be in the reading of an XML file. The solution I've
gleaned so far is that instead of creating another build, I could log in and
"Reload Configuration from Disk"
Or I could just try restarting.
4. This didn't happen on the older server. It won't last forever on this
one.
5. God knows, somebody will come up with a better answer than I can. But
since you're all looking at me at this moment, I say this: "Don't panic. We
still know where our towel is."
Chris
--
Best regards,
Igor Stasenko.
Chris Cunnington
2012-10-23 23:33:28 UTC
Permalink
Post by Igor Stasenko
Hi, Chris..
yes, Jenkins quality remains to be an issue.
We also experiencing problems with it time to time.
I wonder if there any better alternatives..
From other side, jenkins provides a rich set of features (many from plugins),
which would be hard to match if one would start writing it from scratch.
I was just asking my self: "Do they have this problem at INRIA Lille?
They have such a large collection of Jenkins jobs..."
And so, it is nice of you to say that it happens sometimes at your site
as well. Thanks for that.
I figure this stuff will iron itself out over time.

Chris
Craig Latta
2012-10-24 00:24:25 UTC
Permalink
It's time for our podcast recording session on Skype...


-C

--
Craig Latta
www.netjam.org/resume
+31 6 2757 7177
+ 1 415 287 3547 (no SMS)
Frank Shearar
2012-10-24 00:19:26 UTC
Permalink
On 23 October 2012 19:00, Chris Cunnington
Post by Chris Cunnington
I think I can sketch out a few details.
1. The first thing is that this is not a medical emergency. I'm saying that
for my own peace of mind. System administration under duress is no fun at
all.
Ideally one shouldn't do system adminstration when feeling under pressure :)
Post by Chris Cunnington
2. The builds are still there. They don't disappear. They just aren't seen
by the webpage. We are not alone in this. It has happened before.
http://jenkins.361315.n4.nabble.com/JIRA-JENKINS-11938-Jenkins-loses-builds-when-restarted-td4124991.html
3. The problem may be in the reading of an XML file. The solution I've
gleaned so far is that instead of creating another build, I could log in and
"Reload Configuration from Disk"
Or I could just try restarting.
4. This didn't happen on the older server. It won't last forever on this
one.
5. God knows, somebody will come up with a better answer than I can. But
since you're all looking at me at this moment, I say this: "Don't panic. We
still know where our towel is."
Not panicking. At least, not yet :)

frank
Post by Chris Cunnington
Chris
Tom Rushworth
2012-10-24 00:32:50 UTC
Permalink
Hi,

It's reliable enough that I did get a zip of the workspace contents :).

I'll give them a go on OS X server in another 6 hours or so, I'm out of
time right now.

Thanks to both of you for your efforts on our behalf!
Post by Frank Shearar
On 23 October 2012 19:00, Chris Cunnington
Post by Chris Cunnington
I think I can sketch out a few details.
[snip of a bunch of admin discussion]
Post by Frank Shearar
frank
Post by Chris Cunnington
Chris
Regards,
--
Tom Rushworth
Levente Uzonyi
2012-10-24 03:43:43 UTC
Permalink
Isn't the issue simply that Jenkins is looking for the workspace at the
wrong place? When we first set up our Jenkins instance it created the
workspace directory at a different place than what the documentation[1]
says. There was a workspace directory in the JENKINS_HOME directory with
subdirectories for each job IIRC. After the first restart it was looking
for the workspace directory at the place suggested by the documentation,
so our jobs "disappeared". We fixed the directory structure by hand and
everything is fine since then.


Levente

[1] https://wiki.jenkins-ci.org/display/JENKINS/Administering+Jenkins
Post by Chris Cunnington
I think I can sketch out a few details.
1. The first thing is that this is not a medical emergency. I'm saying that for my own peace of mind. System administration under duress is no fun at all.
2. The builds are still there. They don't disappear. They just aren't seen by the webpage. We are not alone in this. It has happened before.
http://jenkins.361315.n4.nabble.com/JIRA-JENKINS-11938-Jenkins-loses-builds-when-restarted-td4124991.html
?
"Reload Configuration from Disk"
Or I could just try restarting.
4. This didn't happen on the older server. It won't last forever on this one.
5. God knows, somebody will come up with a better answer than I can. But since you're all looking at me at this moment, I say this: "Don't panic. We still know where our towel is."
Chris
Chris Cunnington
2012-10-24 04:57:12 UTC
Permalink
Post by Levente Uzonyi
Isn't the issue simply that Jenkins is looking for the workspace at
the wrong place? When we first set up our Jenkins instance it created
the workspace directory at a different place than what the
documentation[1] says. There was a workspace directory in the
JENKINS_HOME directory with subdirectories for each job IIRC. After
the first restart it was looking for the workspace directory at the
place suggested by the documentation, so our jobs "disappeared". We
fixed the directory structure by hand and everything is fine since then.
Levente
[1] https://wiki.jenkins-ci.org/display/JENKINS/Administering+Jenkins
I' bet you're right. I'll look into that.

Chris
Igor Stasenko
2012-10-24 05:50:22 UTC
Permalink
Err, no.
A "workspace" is where slave doing its work.
But artifacts made by job(s) are stored in separate place.

By analogy to real life: you have a workshop where you produce things,
and storage where you put them after.
Post by Levente Uzonyi
Isn't the issue simply that Jenkins is looking for the workspace at the
wrong place? When we first set up our Jenkins instance it created the
workspace directory at a different place than what the documentation[1]
says. There was a workspace directory in the JENKINS_HOME directory with
subdirectories for each job IIRC. After the first restart it was looking for
the workspace directory at the place suggested by the documentation, so our
jobs "disappeared". We fixed the directory structure by hand and everything
is fine since then.
Levente
[1] https://wiki.jenkins-ci.org/display/JENKINS/Administering+Jenkins
Post by Chris Cunnington
I think I can sketch out a few details.
1. The first thing is that this is not a medical emergency. I'm saying
that for my own peace of mind. System administration under duress is no fun
at all.
2. The builds are still there. They don't disappear. They just aren't seen
by the webpage. We are not alone in this. It has happened before.
http://jenkins.361315.n4.nabble.com/JIRA-JENKINS-11938-Jenkins-loses-builds-when-restarted-td4124991.html
3. The problem may be in the reading of an XML file. The solution I've
gleaned so far is that instead of creating another build, I could log in and
"Reload Configuration from Disk"
Or I could just try restarting.
4. This didn't happen on the older server. It won't last forever on this one.
5. God knows, somebody will come up with a better answer than I can. But
since you're all looking at me at this moment, I say this: "Don't panic. We
still know where our towel is."
Chris
--
Best regards,
Igor Stasenko.
Levente Uzonyi
2012-10-24 19:46:09 UTC
Permalink
Post by Igor Stasenko
Err, no.
A "workspace" is where slave doing its work.
But artifacts made by job(s) are stored in separate place.
Right, but as I said, the directory structure created by Jenkins didn't
match what the wiki says when we set up Jenkins. So, tt's still worth
checking the directory structure.


Levente
Post by Igor Stasenko
By analogy to real life: you have a workshop where you produce things,
and storage where you put them after.
Post by Levente Uzonyi
Isn't the issue simply that Jenkins is looking for the workspace at the
wrong place? When we first set up our Jenkins instance it created the
workspace directory at a different place than what the documentation[1]
says. There was a workspace directory in the JENKINS_HOME directory with
subdirectories for each job IIRC. After the first restart it was looking for
the workspace directory at the place suggested by the documentation, so our
jobs "disappeared". We fixed the directory structure by hand and everything
is fine since then.
Levente
[1] https://wiki.jenkins-ci.org/display/JENKINS/Administering+Jenkins
Post by Chris Cunnington
I think I can sketch out a few details.
1. The first thing is that this is not a medical emergency. I'm saying
that for my own peace of mind. System administration under duress is no fun
at all.
2. The builds are still there. They don't disappear. They just aren't seen
by the webpage. We are not alone in this. It has happened before.
http://jenkins.361315.n4.nabble.com/JIRA-JENKINS-11938-Jenkins-loses-builds-when-restarted-td4124991.html
3. The problem may be in the reading of an XML file. The solution I've
gleaned so far is that instead of creating another build, I could log in and
"Reload Configuration from Disk"
Or I could just try restarting.
4. This didn't happen on the older server. It won't last forever on this one.
5. God knows, somebody will come up with a better answer than I can. But
since you're all looking at me at this moment, I say this: "Don't panic. We
still know where our towel is."
Chris
--
Best regards,
Igor Stasenko.
Chris Cunnington
2012-10-24 22:06:38 UTC
Permalink
Post by Levente Uzonyi
Post by Igor Stasenko
Err, no.
A "workspace" is where slave doing its work.
But artifacts made by job(s) are stored in separate place.
Right, but as I said, the directory structure created by Jenkins
didn't match what the wiki says when we set up Jenkins. So, tt's still
worth checking the directory structure.
Levente
The directory structure seems to be as it should be:

xxxxxxxxx:/xxxxxxxxx/jobs/SqueakTrunk/builds/20$ ls -la
total 864
drwxr-xr-x 3 jenkins xxxxxx 4096 Oct 23 19:26 .
drwxr-xr-x 21 jenkins xxxxxx 4096 Oct 23 19:21 ..
drwxr-xr-x 3 jenkins xxxxxx 4096 Oct 23 19:26 archive
-rw-r--r-- 1 jenkins xxxxxx 2906 Oct 23 19:26 build.xml
-rw-r--r-- 1 jenkins xxxxxx 252 Oct 23 19:21 changelog.xml
-rw-r--r-- 1 jenkins xxxxxx 857130 Oct 23 19:26 junitResult.xml
-rw-r--r-- 1 jenkins xxxxxx 1312 Oct 23 19:26 log

The deliverables are in the archive directory as expected. I think the
build.xml is what is read to inform the web page.

Looking at the site it seems a bit more stable than yesterday. Somewhat.
Nothing has disappeared so far. Perhaps it's hardening a bit with use.

Chris
Frank Shearar
2012-10-24 22:09:53 UTC
Permalink
On 24 October 2012 18:06, Chris Cunnington
Post by Chris Cunnington
Post by Levente Uzonyi
Post by Igor Stasenko
Err, no.
A "workspace" is where slave doing its work.
But artifacts made by job(s) are stored in separate place.
Right, but as I said, the directory structure created by Jenkins didn't
match what the wiki says when we set up Jenkins. So, tt's still worth
checking the directory structure.
Levente
xxxxxxxxx:/xxxxxxxxx/jobs/SqueakTrunk/builds/20$ ls -la
total 864
drwxr-xr-x 3 jenkins xxxxxx 4096 Oct 23 19:26 .
drwxr-xr-x 21 jenkins xxxxxx 4096 Oct 23 19:21 ..
drwxr-xr-x 3 jenkins xxxxxx 4096 Oct 23 19:26 archive
-rw-r--r-- 1 jenkins xxxxxx 2906 Oct 23 19:26 build.xml
-rw-r--r-- 1 jenkins xxxxxx 252 Oct 23 19:21 changelog.xml
-rw-r--r-- 1 jenkins xxxxxx 857130 Oct 23 19:26 junitResult.xml
-rw-r--r-- 1 jenkins xxxxxx 1312 Oct 23 19:26 log
The deliverables are in the archive directory as expected. I think the
build.xml is what is read to inform the web page.
Looking at the site it seems a bit more stable than yesterday. Somewhat.
Nothing has disappeared so far. Perhaps it's hardening a bit with use.
The glue's setting?

frank
Post by Chris Cunnington
Chris
Chris Cunnington
2012-10-24 22:11:58 UTC
Permalink
Post by Frank Shearar
On 24 October 2012 18:06, Chris Cunnington
Post by Chris Cunnington
Post by Levente Uzonyi
Post by Igor Stasenko
Err, no.
A "workspace" is where slave doing its work.
But artifacts made by job(s) are stored in separate place.
Right, but as I said, the directory structure created by Jenkins didn't
match what the wiki says when we set up Jenkins. So, tt's still worth
checking the directory structure.
Levente
xxxxxxxxx:/xxxxxxxxx/jobs/SqueakTrunk/builds/20$ ls -la
total 864
drwxr-xr-x 3 jenkins xxxxxx 4096 Oct 23 19:26 .
drwxr-xr-x 21 jenkins xxxxxx 4096 Oct 23 19:21 ..
drwxr-xr-x 3 jenkins xxxxxx 4096 Oct 23 19:26 archive
-rw-r--r-- 1 jenkins xxxxxx 2906 Oct 23 19:26 build.xml
-rw-r--r-- 1 jenkins xxxxxx 252 Oct 23 19:21 changelog.xml
-rw-r--r-- 1 jenkins xxxxxx 857130 Oct 23 19:26 junitResult.xml
-rw-r--r-- 1 jenkins xxxxxx 1312 Oct 23 19:26 log
The deliverables are in the archive directory as expected. I think the
build.xml is what is read to inform the web page.
Looking at the site it seems a bit more stable than yesterday. Somewhat.
Nothing has disappeared so far. Perhaps it's hardening a bit with use.
The glue's setting?
frank
The epoxy. Yes.

Chris
Frank Shearar
2012-10-24 02:19:22 UTC
Permalink
On 23 October 2012 18:21, Chris Cunnington
Post by Chris Cunnington
Post by Frank Shearar
Post by Tom Rushworth
Hi Frank,
I made a stab at getting the new image to run on my OS X server, but the
link below doesn't seem to do anything in a regular browser, or when I
paste it into the update-image.st script as the update source.
I'm probably completely misunderstanding what I should be doing...
Can you give me a bigger hint? :).
Right. This is the second time we've seen this. There's _supposed_ to
be a zip there for you to download.
Chris Cunnington (I'm not saying that with my Dad Voice; there are
just lots of people sharing your name on the list), did you do
anything to the build server? It looks like all the build history for
SqueakTrunk just vanished.
frank
http://www.squeakci.org
Yay!

frank
Post by Chris Cunnington
Chris
Herbert König
2012-10-24 01:29:55 UTC
Permalink
Hi,
Post by Frank Shearar
We're down to 6 failing unit tests on the Linux platform, one of which
is an expected failure and another of which is a known network issue
for the Linux platform.
Would my kind Windows and OSX build volunteers please run builds
against the latest trunk image?
(http://173.246.101.237:8080/job/SqueakTrunk/lastCompletedBuild/artifact/)
Yesterday (Oct 22) I could download build #17 and got a TrunkImage.image
and .changes. Not the usual zip including the VM.

This is update 12251 according to the system reporter. I disable IPV6
support via preferences and run all tests in the Cog VM I used for the
previous tests. See http://minus.com/luZyqW2PRhIrq.

I get 3222 run, 3178 passes, 18 expected failures, 5 failures, 21
errors, 0 unexpected passes, as if the network preference is ignored again.
See http://minus.com/lpibYAKEi3bw5
I did this twice, as I didn't want to believe it.

If I update this image to trunk, it loads some updates and again claims
to be 12251 version.
Tests like before result in 3219 run, 3195 passes, 18 expected failures,
5 failures, 1 errors, 0 unexpected passes. See:
http://minus.com/l17FSZY5yOnhH

I have no idea but I still have that image if someone wants it.

http://minus.com/mbam1m6GnSJJJH is the complete Folder with the
screenshots of all runs I published up to now.

Cheers

Herbert
Frank Shearar
2012-10-24 17:16:14 UTC
Permalink
Hi,
Post by Frank Shearar
We're down to 6 failing unit tests on the Linux platform, one of which
is an expected failure and another of which is a known network issue
for the Linux platform.
Would my kind Windows and OSX build volunteers please run builds
against the latest trunk image?
(http://173.246.101.237:8080/job/SqueakTrunk/lastCompletedBuild/artifact/)
Yesterday (Oct 22) I could download build #17 and got a TrunkImage.image and
.changes. Not the usual zip including the VM.
The VM comes from the repository at
https://github.com/frankshearar/squeak-ci. In particular, you can get
a copy directly from here:
https://github.com/frankshearar/squeak-ci/zipball/master

Sorry, I thought I'd only need to mention the image location.
This is update 12251 according to the system reporter. I disable IPV6
support via preferences and run all tests in the Cog VM I used for the
previous tests. See http://minus.com/luZyqW2PRhIrq.
I get 3222 run, 3178 passes, 18 expected failures, 5 failures, 21 errors, 0
unexpected passes, as if the network preference is ignored again.
See http://minus.com/lpibYAKEi3bw5
I did this twice, as I didn't want to believe it.
If I update this image to trunk, it loads some updates and again claims to
be 12251 version.
Tests like before result in 3219 run, 3195 passes, 18 expected failures, 5
http://minus.com/l17FSZY5yOnhH
I have no idea but I still have that image if someone wants it.
http://minus.com/mbam1m6GnSJJJH is the complete Folder with the screenshots
of all runs I published up to now.
Thanks very much for running the tests, Herbert. I appreciate it!

This is actually what I _expected_, since one of the 6 failing tests
is a network issue on the Linux platform. That is, I expect
NetworkTests.Kernel.SocketTest.testSendTimeout to _pass_ on Windows.

That error is news to me though: FileDirectoryTest.testRelativeNameIfAbsolut...

Would you mind sending more details on that failed test? I'd like to
see which assertion fails.

Thanks again!

frank
Cheers
Herbert
Levente Uzonyi
2012-10-24 19:14:44 UTC
Permalink
Post by Frank Shearar
Hi,
Post by Frank Shearar
We're down to 6 failing unit tests on the Linux platform, one of which
is an expected failure and another of which is a known network issue
for the Linux platform.
Would my kind Windows and OSX build volunteers please run builds
against the latest trunk image?
(http://173.246.101.237:8080/job/SqueakTrunk/lastCompletedBuild/artifact/)
Yesterday (Oct 22) I could download build #17 and got a TrunkImage.image and
.changes. Not the usual zip including the VM.
The VM comes from the repository at
https://github.com/frankshearar/squeak-ci. In particular, you can get
https://github.com/frankshearar/squeak-ci/zipball/master
Sorry, I thought I'd only need to mention the image location.
This is update 12251 according to the system reporter. I disable IPV6
support via preferences and run all tests in the Cog VM I used for the
previous tests. See http://minus.com/luZyqW2PRhIrq.
I get 3222 run, 3178 passes, 18 expected failures, 5 failures, 21 errors, 0
unexpected passes, as if the network preference is ignored again.
See http://minus.com/lpibYAKEi3bw5
I did this twice, as I didn't want to believe it.
If I update this image to trunk, it loads some updates and again claims to
be 12251 version.
Tests like before result in 3219 run, 3195 passes, 18 expected failures, 5
http://minus.com/l17FSZY5yOnhH
I have no idea but I still have that image if someone wants it.
http://minus.com/mbam1m6GnSJJJH is the complete Folder with the screenshots
of all runs I published up to now.
Thanks very much for running the tests, Herbert. I appreciate it!
This is actually what I _expected_, since one of the 6 failing tests
is a network issue on the Linux platform. That is, I expect
NetworkTests.Kernel.SocketTest.testSendTimeout to _pass_ on Windows.
It randomly passes on Vista, but fails most of the time. The problem is
that the test expects that if it fills up the send buffer of a socket,
then all subsequent send requests will fail, but I think the OS somehow
makes the send buffer larger, so while sending failed at a point in time,
later it will work again.
Post by Frank Shearar
That error is news to me though: FileDirectoryTest.testRelativeNameIfAbsolut...
Would you mind sending more details on that failed test? I'd like to
see which assertion fails.
That's an "old" error on Windows. The method (#relativeNameIfAbsoluteFor:)
expects that an absolute path begins with the path delimitar, which is
obviously not true on windows (C:\... doesn't begin with \). There are
several absolute path formats on windows, so I think the method has to be
implementation in DosFileDirectory.


Levente
Post by Frank Shearar
Thanks again!
frank
Cheers
Herbert
Frank Shearar
2012-10-24 22:35:19 UTC
Permalink
Post by Frank Shearar
We're down to 6 failing unit tests on the Linux platform, one of which
is an expected failure and another of which is a known network issue
for the Linux platform.
Would my kind Windows and OSX build volunteers please run builds
against the latest trunk image?
(http://173.246.101.237:8080/job/SqueakTrunk/lastCompletedBuild/artifact/)
Note the new URL: that's thanks to Chris Cunnington, who has kindly
moved Jenkins to its new home. That will no doubt shortly become the
new squeakci.org.
Windows users note please that you'll need to grant permission for the
build executables etc to act as servers and to make connections: you
can do this through the Windows Firewall configuration screens. (I'd
give more detailed instructions, but my Windows machines are far away
right now.)
(Dale, this shouldn't affect you much, in the sense that you won't
need any changes to builderCI or whatever once the DNS changes have
been made. So if they _do_ affect you, they shouldn't affect you for
very long.)
It's about time I started with the more admin-y side of the release,
making a Changelog and such. I'll make a start on this soon,
docs/progress to be on http://wiki.squeak.org/squeak/6188
frank
The latest builds have a few new failures:

* Tests.Compiler.DecompilerTests.testDecompilerInClassesLAtoLM [1]
* TraitsTests.Kernel.ClassTraitTest.testConflictsAliasesAndExclusions [2]

which is kind've surprising.

frank

[1] Error Message

subscript is out of bounds: 1
Stacktrace

Array(Object)>>error:
Array(Object)>>errorSubscriptBounds:
Array(Object)>>at:
Decompiler>>pushTemporaryVariable:
Decompiler(InstructionStream)>>interpretNextInstructionFor:
Decompiler>>interpretNextInstructionFor:
Decompiler>>statementsTo:
Decompiler>>blockTo:
Decompiler>>doClosureCopyCopiedValues:numArgs:blockSize:
Decompiler>>pushClosureCopyNumCopiedValues:numArgs:blockSize:
Decompiler(InstructionStream)>>interpretExtension:in:for:
Decompiler(InstructionStream)>>interpretNextInstructionFor:
Decompiler>>interpretNextInstructionFor:
Decompiler>>statementsTo:
Decompiler>>blockTo:
Decompiler>>decompile:in:method:using:
Decompiler>>decompile:in:method:
DecompilerTests>>checkDecompileMethod:
[] in [] in [] in DecompilerTests>>decompileClassesSelect:
MethodDictionary>>keysAndValuesDo:
LargeNegativeIntegerTest class(Behavior)>>selectorsAndMethodsDo:
[] in [] in DecompilerTests>>decompileClassesSelect:
Array(SequenceableCollection)>>do:
[] in DecompilerTests>>decompileClassesSelect:
BlockClosure>>on:do:
CurrentReadOnlySourceFiles class>>cacheDuring:
DecompilerTests>>decompileClassesSelect:
DecompilerTests>>testDecompilerInClassesLAtoLM
DecompilerTests(TestCase)>>performTest


[2] Stacktrace

Parser>>notify:at:
Parser>>expected:
Parser>>pattern:inContext:
Parser>>method:context:
[] in Parser>>parse:class:category:noPattern:context:notifying:ifFail:
BlockClosure>>on:do:
Parser>>parse:class:category:noPattern:context:notifying:ifFail:
Compiler>>translate:noPattern:ifFail:
Compiler>>compile:in:classified:notifying:ifFail:
Metaclass(ClassDescription)>>traitAddSelector:withMethod:
[] in Metaclass(ClassDescription)>>installTraitMethodDict:
[] in Dictionary>>keysAndValuesDo:
Dictionary>>associationsDo:
Dictionary>>keysAndValuesDo:
Metaclass(ClassDescription)>>installTraitMethodDict:
Metaclass(ClassDescription)>>installTraitsFrom:
Metaclass(ClassDescription)>>updateTraits
Metaclass(ClassDescription)>>classify:under:from:trait:
[] in ClassTrait(TraitDescription)>>notifyOfRecategorizedSelector:from:to:
Array(SequenceableCollection)>>do:
ClassTrait(TraitDescription)>>notifyOfRecategorizedSelector:from:to:
TraitOrganizer(ClassOrganizer)>>notifyOfChangedSelector:from:to:
TraitOrganizer(ClassOrganizer)>>classify:under:suppressIfDefault:logged:
TraitOrganizer(ClassOrganizer)>>classify:under:suppressIfDefault:
TraitOrganizer(Categorizer)>>classify:under:
[] in ClassTrait(ClassDescription)>>installTraitMethodDict:
[] in Dictionary>>keysAndValuesDo:
Dictionary>>associationsDo:
Dictionary>>keysAndValuesDo:
ClassTrait(ClassDescription)>>installTraitMethodDict:
ClassTrait(ClassDescription)>>installTraitsFrom:
ClassTrait(TraitDescription)>>installTraitsFrom:
ClassTrait(ClassDescription)>>updateTraits
[] in ClassTrait(TraitDescription)>>addSelectorSilently:withMethod:
Array(SequenceableCollection)>>do:
ClassTrait(TraitDescription)>>addSelectorSilently:withMethod:
ClassTrait(ClassDescription)>>addAndClassifySelector:withMethod:inProtocol:notifying:
ClassTrait(ClassDescription)>>compile:classified:withStamp:notifying:logSource:
ClassTrait(ClassDescription)>>compile:classified:withStamp:notifying:
ClassTrait(ClassDescription)>>compile:classified:notifying:
ClassTrait(ClassDescription)>>compile:classified:
ClassTraitTest>>testConflictsAliasesAndExclusions
ClassTraitTest(TestCase)>>performTest
Jeff Gonis
2012-10-27 05:21:32 UTC
Permalink
I know that this is quite late, but better late than never. Running the
latest from Frank's squeak-ci repo I get 8 errors:

KernelTests: 3 errors 0 failures.
NetworkTests: 2 errors 0 failures.
Tests: 2 errors 0 failures.
ToolsTests: 1 errors 0 failures.

DependencyBrowserTest>>#testClassList
LargeNegativeIntegerTest>>#testMinimumNegativeArithmetic
LocaleTest>>#testLocaleChanged
ReleaseTest>>#testNoObsoleteClasses
SocketTest>>#testSocketReuse
SocketTest>>#testUDP
TimeStampTest>>#testFromSeconds
TimeStampTest>>#testReadFromA1

I'll run it on my linux vm momentarily and report the results.

Thanks for all the great work Frank!
Jeff
Post by Frank Shearar
Post by Frank Shearar
We're down to 6 failing unit tests on the Linux platform, one of which
is an expected failure and another of which is a known network issue
for the Linux platform.
Would my kind Windows and OSX build volunteers please run builds
against the latest trunk image?
(
http://173.246.101.237:8080/job/SqueakTrunk/lastCompletedBuild/artifact/)
Post by Frank Shearar
Note the new URL: that's thanks to Chris Cunnington, who has kindly
moved Jenkins to its new home. That will no doubt shortly become the
new squeakci.org.
Windows users note please that you'll need to grant permission for the
build executables etc to act as servers and to make connections: you
can do this through the Windows Firewall configuration screens. (I'd
give more detailed instructions, but my Windows machines are far away
right now.)
(Dale, this shouldn't affect you much, in the sense that you won't
need any changes to builderCI or whatever once the DNS changes have
been made. So if they _do_ affect you, they shouldn't affect you for
very long.)
It's about time I started with the more admin-y side of the release,
making a Changelog and such. I'll make a start on this soon,
docs/progress to be on http://wiki.squeak.org/squeak/6188
frank
* Tests.Compiler.DecompilerTests.testDecompilerInClassesLAtoLM [1]
* TraitsTests.Kernel.ClassTraitTest.testConflictsAliasesAndExclusions [2]
which is kind've surprising.
frank
[1] Error Message
subscript is out of bounds: 1
Stacktrace
DecompilerTests>>testDecompilerInClassesLAtoLM
DecompilerTests(TestCase)>>performTest
[2] Stacktrace
Metaclass(ClassDescription)>>updateTraits
ClassTrait(ClassDescription)>>updateTraits
ClassTraitTest>>testConflictsAliasesAndExclusions
ClassTraitTest(TestCase)>>performTest
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20121026/77d0c582/attachment.htm
Jeff Gonis
2012-10-27 08:46:33 UTC
Permalink
So running it on my ubuntu 12.10 linux vm resulted in 10 failures.

The additional failures appeared to be Sockets.

DependencyBrowserTest>>#testClassList
LargeNegativeIntegerTest>>#testMinimumNegativeIntegerArithmetic
LocaleTest>>#testLocaleChanged
ReleaseTest>>#testNoObsoleteClasses
SocketTest>>#testLocalAddress
SocketTest>>#testPeerName
SocketTest>>#testRemoteAddress
SocketTest>>#testSendTimeout
TimeStampTest>>#testFromSeconds
TimeStampTest>>#testReadFromA1

I'll run these tests on each system sometime again tomorrow.
Thanks,
Jeff
Post by Jeff Gonis
I know that this is quite late, but better late than never. Running the
KernelTests: 3 errors 0 failures.
NetworkTests: 2 errors 0 failures.
Tests: 2 errors 0 failures.
ToolsTests: 1 errors 0 failures.
DependencyBrowserTest>>#testClassList
LargeNegativeIntegerTest>>#testMinimumNegativeArithmetic
LocaleTest>>#testLocaleChanged
ReleaseTest>>#testNoObsoleteClasses
SocketTest>>#testSocketReuse
SocketTest>>#testUDP
TimeStampTest>>#testFromSeconds
TimeStampTest>>#testReadFromA1
I'll run it on my linux vm momentarily and report the results.
Thanks for all the great work Frank!
Jeff
Post by Frank Shearar
Post by Frank Shearar
We're down to 6 failing unit tests on the Linux platform, one of which
is an expected failure and another of which is a known network issue
for the Linux platform.
Would my kind Windows and OSX build volunteers please run builds
against the latest trunk image?
(
http://173.246.101.237:8080/job/SqueakTrunk/lastCompletedBuild/artifact/)
Post by Frank Shearar
Note the new URL: that's thanks to Chris Cunnington, who has kindly
moved Jenkins to its new home. That will no doubt shortly become the
new squeakci.org.
Windows users note please that you'll need to grant permission for the
build executables etc to act as servers and to make connections: you
can do this through the Windows Firewall configuration screens. (I'd
give more detailed instructions, but my Windows machines are far away
right now.)
(Dale, this shouldn't affect you much, in the sense that you won't
need any changes to builderCI or whatever once the DNS changes have
been made. So if they _do_ affect you, they shouldn't affect you for
very long.)
It's about time I started with the more admin-y side of the release,
making a Changelog and such. I'll make a start on this soon,
docs/progress to be on http://wiki.squeak.org/squeak/6188
frank
* Tests.Compiler.DecompilerTests.testDecompilerInClassesLAtoLM [1]
* TraitsTests.Kernel.ClassTraitTest.testConflictsAliasesAndExclusions [2]
which is kind've surprising.
frank
[1] Error Message
subscript is out of bounds: 1
Stacktrace
DecompilerTests>>testDecompilerInClassesLAtoLM
DecompilerTests(TestCase)>>performTest
[2] Stacktrace
Metaclass(ClassDescription)>>updateTraits
ClassTrait(ClassDescription)>>updateTraits
ClassTraitTest>>testConflictsAliasesAndExclusions
ClassTraitTest(TestCase)>>performTest
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20121026/db2207f0/attachment.htm
Frank Shearar
2012-10-27 13:26:02 UTC
Permalink
Post by Jeff Gonis
I know that this is quite late, but better late than never. Running the
KernelTests: 3 errors 0 failures.
NetworkTests: 2 errors 0 failures.
Tests: 2 errors 0 failures.
ToolsTests: 1 errors 0 failures.
DependencyBrowserTest>>#testClassList
I also saw this guy and was a bit surprised. I mean, the test is
fragile in that we examine the current methods of a few classes
(String) - I wrote it that way for lack of mocks - but didn't have a
chance to see why. It probably broke because of the recent String
changes shuffled the method order.
Post by Jeff Gonis
LargeNegativeIntegerTest>>#testMinimumNegativeArithmetic
General question not just for Jeff: Can we mark this as an expected failure?

frank
Post by Jeff Gonis
LocaleTest>>#testLocaleChanged
ReleaseTest>>#testNoObsoleteClasses
SocketTest>>#testSocketReuse
SocketTest>>#testUDP
TimeStampTest>>#testFromSeconds
TimeStampTest>>#testReadFromA1
I'll run it on my linux vm momentarily and report the results.
Thanks for all the great work Frank!
Jeff
Post by Frank Shearar
Post by Frank Shearar
We're down to 6 failing unit tests on the Linux platform, one of which
is an expected failure and another of which is a known network issue
for the Linux platform.
Would my kind Windows and OSX build volunteers please run builds
against the latest trunk image?
(http://173.246.101.237:8080/job/SqueakTrunk/lastCompletedBuild/artifact/)
Note the new URL: that's thanks to Chris Cunnington, who has kindly
moved Jenkins to its new home. That will no doubt shortly become the
new squeakci.org.
Windows users note please that you'll need to grant permission for the
build executables etc to act as servers and to make connections: you
can do this through the Windows Firewall configuration screens. (I'd
give more detailed instructions, but my Windows machines are far away
right now.)
(Dale, this shouldn't affect you much, in the sense that you won't
need any changes to builderCI or whatever once the DNS changes have
been made. So if they _do_ affect you, they shouldn't affect you for
very long.)
It's about time I started with the more admin-y side of the release,
making a Changelog and such. I'll make a start on this soon,
docs/progress to be on http://wiki.squeak.org/squeak/6188
frank
* Tests.Compiler.DecompilerTests.testDecompilerInClassesLAtoLM [1]
* TraitsTests.Kernel.ClassTraitTest.testConflictsAliasesAndExclusions [2]
which is kind've surprising.
frank
[1] Error Message
subscript is out of bounds: 1
Stacktrace
DecompilerTests>>testDecompilerInClassesLAtoLM
DecompilerTests(TestCase)>>performTest
[2] Stacktrace
Metaclass(ClassDescription)>>updateTraits
ClassTrait(ClassDescription)>>updateTraits
ClassTraitTest>>testConflictsAliasesAndExclusions
ClassTraitTest(TestCase)>>performTest
Continue reading on narkive:
Loading...