Martin v. Löwis added the comment:
I think I fixed it. Please try again, and report whether it works.
--
nosy: +loewis
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/meta/issue
New submission from Martin v. Löwis:
Please add a reverse DNS mapping for 2a01:4f8:131:2480::3 pointing to
bugs.python.org. Apparently, otherwise email is considered as spam; see issue562
--
assignedto: izak
messages: 2969
nosy: izak, loewis, roche
priority: urgent
status: unread
Martin v. Löwis added the comment:
Starting to implement an OpenID Connect Consumer using the oic library, I fail
in the first step. The OpenID specification says in
http://openid.net/specs/openid-connect-discovery-1_0.html#ProviderMetadata
that the issuer field of the metadata MUST be an
Martin v. Löwis added the comment:
As of python-dev:51dd28d02f07, we now send openid_shutdown_ack in requests to
Google; this should suppress the reauthentication.
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/meta/issue
Martin v. Löwis added the comment:
Started looking into it.
--
status: unread -> chatting
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/meta/is
Am 19.09.14 17:19, schrieb Benjamin Peterson:
> +tracker-discuss where people with access more likely reside.
>
> On Fri, Sep 19, 2014, at 06:23, Donald Stufft wrote:
>> Hey,
>>
>> Is it possible for me to get access to the current server hosting
>> bugs.p.o? Is that a possibility?
I have now add
Martin v. Löwis added the comment:
I cannot reproduce the issue. On http://bugs.python.org/issue2771, I just
changed the value of "Status" successfully to "no selection". I got a green bar
saying that the change was successful, and the issue log shows that the
Martin v. Löwis added the comment:
The "official" way seems to be the SQL shell of django, so I did
roundup@psf:~/trackers/tracker/rietveld$ python manage.py shell
>>> from django.contrib.sites.models import Site
>>> Site.objects.create(name='bugs.python.or
Martin v. Löwis added the comment:
I have now moved the notice to the page level.
--
nosy: +loewis
status: chatting -> resolved
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/meta/is
Martin v. Löwis added the comment:
It's actually simpler than that. Roundup reports the value of _file._type for
Content-type when a file is downloaded. This is what the auditor sets, but it
remains unchanged for the already-uploaded files. I have now changed it for all
existing files i
Martin v. Löwis added the comment:
Why did you want to change the status to "- no selection -"?
--
nosy: +loewis
status: unread -> chatting
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/round
Martin v. Löwis added the comment:
I have now removed it from the tracker.
--
status: chatting -> resolved
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/meta/is
Martin v. Löwis added the comment:
I have now removed the notices about SourceForge.
As for spam accounts: it doesn't appear to be an urgent issue (remaining
untreated for a year). There is spam protection already in place, preventing
registration from bots who fill out the form too qu
Martin v. Löwis added the comment:
If somebody want to test this, I suggest that we should pull the upstream
source, merge with our own branch, deploy this, test it, then revert to the
version that we had deployed before (hoping that there won't be any schema
upgrades in be
Martin v. Löwis added the comment:
Yes, it can be closed
--
status: chatting -> resolved
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/meta/is
Martin v. Löwis added the comment:
I don't think that the spam risk is high, so I think it might still be
desirable to have it. Given that I had no time to work on this for the last two
years, chances are that I won't have time in the coming two yea
Martin v. Löwis added the comment:
This is almost the right place to bring this up; please use tracker-discuss
(mail.python.org) instead.
The subject was indeed discussed and dismissed long ago. We set up a working
group in 2005 to pick a new bug tracker for Python, moving away from
Martin v. Löwis added the comment:
I have changed the type of all html files in this tracker to text/plain.
I'll also look into turning off the setuptools tracker; I doubt it is active
still.
--
nosy: +loewis
___
PSF Meta Tr
Martin v. Löwis added the comment:
I have now installed a CACert certificate, so https is available for
bugs.python.org
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/meta/issue
Martin v. Löwis added the comment:
I think that the version should stay retired. If it was available, people would
continue to report issues specifically against this version, and it would take
maintenance time to clarify that the issues will not get fixed. I believe this
avoidable overhead
Martin v. Löwis added the comment:
Now that Izak has restored connectivity, I have restored the DNS entry.
--
status: chatting -> resolved
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/meta/is
Martin v. Löwis added the comment:
techtonik specifically asked for FreeSSL, so I still wonder what that is.
I'm familiar with StartSSL, and the PSF infrastructure group indeed does have
access to such certificates. The problem with the StartSSL certificate is that
it has "python
Martin v. Löwis added the comment:
What is FreeSSL, and how do I get them to issue a free certificate?
--
nosy: +loewis
status: unread -> chatting
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/meta/is
Martin v. Löwis added the comment:
iwontbecreative: your conclusion is incorrect. The database doesn't store plain
text passwords, but hashes, so it sorts on hash. With that, it is not possible
to recover a user's password.
--
nosy: +loewis
status: unread -
Martin v. Löwis added the comment:
iwontbecreative: your conclusion is incorrect. The database doesn't store plain
text passwords, but hashes, so it sorts on hash. With that, it is not possible
to recover a user's password.
--
nos
Martin v. Löwis added the comment:
I can't reproduce this in the reverse direction; pinging 195.5.228.255 works
fine. If the problem persists, please report your source IP.
For the record, here is the reverse trace
traceroute to 195.5.228.255 (195.5.228.255), 30 hops max, 60 byte packe
Martin v. Löwis added the comment:
follow-up: apparently, there are browser-specific plugins (like "OpenID for
Firefox" or "OpenID Autofill" for Chrome) which you may want to used.
As for having to relogin: I cannot reproduce this. With Chrome, the session
happily surviv
Martin v. Löwis added the comment:
I fail to see the issue. Whether or not the browser saves the form value is a
browser issue (and possible a user issue, for configuring the browser
appropriately), not one of Roundup.
The OpenID specification defines that Roundup should give a hint, by
Martin v. Löwis added the comment:
Exactly that.
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/meta/issue503>
___
___
T
Martin v. Löwis added the comment:
The document returned from the URL is a text/plain document. As per OpenID
spec, it ust be a HTML document, see
http://openid.net/specs/openid-authentication-2_0.html#html_disco
The implementation recognizes HTML documents from the MIME types text/html and
As you know, wiki.python.org was broken into, which means that
there is a risk that the passwords were stolen (although it's
uncertain whether that actually happened).
There is a chance that some users were using the same passwords
on different services, which caused PyPI maintainers to force
a p
Am 17.02.13 20:41, schrieb fwierzbi...@gmail.com:
Hi all,
Having played around with a local tracker, I think I should be able to
fix Jython's version problems with the roundup admin tool. How do I
access the Jython tracker instance? Has it been moved to the VM
infrastructure?
No. Your account
Martin v. Löwis added the comment:
-1 on the entire issue - no action should be taken.
The real problem is that there is a bug in roundup in each and every case where
the email is sent. So the proper action is to fix the bug, not to massage the
error message. python-dev readers should easily
Martin v. Löwis added the comment:
storchaka: is deleting multiple files in one action really desirable? If so,
create a separate issue: this issue is about attaching multiple files.
--
nosy: +loewis
___
PSF Meta Tracker
<h
Martin v. Löwis added the comment:
I don't know; I don't think there is a tracker for it. You should probably ask
the maintainers of the hg installation (i.e. Antoine and Georg).
--
status: chatting -> resolved
___
PSF
Martin v. Löwis added the comment:
This is not an issue of the bug tracker (i.e. bugs.python.org).
--
nosy: +loewis
status: chatting -> resolved
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/meta/is
Martin v. Löwis added the comment:
I agree that the icon shouldn't be displayed in every place the name occurs.
I'm skeptical whether it has value in the user search list. If the complaint is
only about consistency, I propose to make it consistent by also dropping the
"contri
Martin v. Löwis added the comment:
Commiter -> Committer
Regards,
Martin
--
title: Cannot edit (eg) component: No such action "editcsv" -> Typo
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/ro
Martin v. Löwis added the comment:
It was temporarily disabled in r70258 (Mar 9 06:19:15 2009)
http://svn.python.org/view/tracker/roundup-src/roundup/cgi/client.py?r1=70258&r2=70388&pathrev=88900
but re-enabled in r70388 (Mar 15 21:43:30 2009)
http://svn.python.org/view/tracker/rou
Martin v. Löwis added the comment:
Indeed.
barry: can you please report the exact URL your browser was pointing to when
you entered the number? Performing your instructions fails to reproduce the
issue; entering 4194 after typing "bugs.python.org" in the URL bar in Chrome
works jus
Martin v. Löwis added the comment:
One issue at a time, please. Your original request was to get more information
than from CSV export, which I think is satisfied with the XML-RPC access, so
I'm closing this issue as resolved.
Notice that the Python bug tracker really is an installati
Martin v. Löwis added the comment:
This issue has a long history. Originally, search was only for open issues.
Then, on request by many users, it was changed to search for all issues
instead, since people would often re-report an issue that was already closed.
So I'm rejecting this re
Martin v. Löwis added the comment:
Ezio: I think this is fine (except perhaps that developers may be unaware what
exactly the ham/spam classification means). To test it, you'll find plenty
messages in the tracker. To change your roles, just use roundup-admin on the
command line (I
Martin v. Löwis added the comment:
This should now be fixed (actually, a work around is in place that requires the
user to complete the registration within 1000s).
--
nosy: +loewis
status: unread -> resolved
___
PSF Meta Trac
Martin v. Löwis added the comment:
Microsoft has now responded saying that they can't find anything off-hand, and
have requested further information, which I have provided.
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roun
Martin v. Löwis added the comment:
There is a problem delivering email to hotmail.com. We have three messages to
this email address pending in the outgoing queue; postfix says "Connection
timed out". A plain telnet to mx2.hotmail.com gives a TCP timeout. I have filed
a support re
Martin v. Löwis added the comment:
Ok, to focus the bug: the issue appears that people cannot log in when a search
results page is displayed. Login seems to work fine otherwise.
--
title: Login not working (mostly) -> Cannot log into search results p
Martin v. Löwis added the comment:
dannystaple: are you sure the issue you encountered is the very same issue as
the one discussed here?
What is the exact sequence of URLs that you navigated to, and what is the error
message that you got? roundup most certainly has a came_from parameter in
Martin v. Löwis added the comment:
The critical thing really is that the spam was accumulating over quite some
time and nobody was looking into it. This, in turn, is because the Jython
tracker is (was?) unmaintained. I have now despammed it, and requested Google
to rescan it, so Google has
Martin v. Löwis added the comment:
Comparing the password hashes is inconclusive; the passwords are salted.
In any case, this issue is about a problem that you perceive for yourself.
Whether or not other people feel likewise threatened, we cannot know.
--
priority: critical -> w
Martin v. Löwis added the comment:
The risk isn't really high. Just chose a password that you don't use anywhere
else, and the threat of somebody stealing it can be safely ignored. Somebody
might be posting in your name, but that doesn't scare me at all.
--
nosy:
Martin v. Löwis added the comment:
It seems that the key issue is that people want to avoid commenting on a closed
issue. I think this is best achieved by hiding any edit UI first, only enabling
editing on explicit request. I suggest to put a text like
"This issue is closed. If you h
Martin v. Löwis added the comment:
> But does this mean that other consumers are not so strict and allow cheating?
I'm not sure how serious this is, but the OpenID spec
(http://openid.net/specs/openid-authentication-2_0.html) seems to say in
section 11 that this MUST be verifie
Martin v. Löwis added the comment:
I see. You need to change your "openid.server" from
"http://openid.yandex.ru/server.xml"; to "http://openid.yandex.ru/server/";,
since this is what openid.yandex.ru says is the official endpoint URL.
The Python bug tracker not
Martin v. Löwis added the comment:
Can you please report what the op_endpoint value is in the redirect back to the
bugtracker? It will be part of the URL, following the string
"&openid.op_endpoint="
Supposedly, this is incorrect, indicating that some cheating happened.
Deleg
Martin v. Löwis added the comment:
Good idea!
--
status: unread -> chatting
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/meta/is
Martin v. Löwis added the comment:
rmtew: search is definitely case insensitive. I cannot reproduce your report;
the search for "UDP" yields plenty results.
The issue here is that the search only looks for whole words indeed, with words
being defined by r'(?u)\b\w{%d
Martin v. Löwis added the comment:
I have now edited /etc/apache2/ports.conf, /etc/ssh/sshd_config, and
/etc/postfix/main.cf to enable IPv6.
For Postfix, I would appreciate a review of the changes; I added inet_protcols
and edited inet_interfaces and mynetworks.
I also put the name into DNS
Martin v. Löwis added the comment:
I'm declaring this resolved (no change), with the rationale given by Daniel
(ajaksu2). If users really don't want their emails published, we should disable
them altogether (option 1). However, users apparently are not strongly
requesting this, plu
Martin v. Löwis added the comment:
This is now fixed in r88949.
--
status: chatting -> resolved
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/meta/is
Martin v. Löwis added the comment:
Unfortunately, the roundup tracker itself is unmaintained, ever since Stefan
Seefeld left. I have now reindexed the setuptools tracker, and the search for
"xapian" does bring up some issues.
--
status: resolved -
Martin v. Löwis added the comment:
This is now fixed in r88948.
--
nosy: +loewis
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/meta/issue
Martin v. Löwis added the comment:
The date specification should be better documented, though: either by
reference, or by copying from
http://roundup.sourceforge.net/doc-1.0/user_guide.html#date-properties
Your examples are specified as "from -5d", "from -1m", and "f
Martin v. Löwis added the comment:
> Why is 'action=logout' in the url? It has already happened
Wrong. The URL you are seeing is the very one that did the
logout. This is how web browsers work: they show you the URL
that corresponds to the page you are seeing. Likewise f
Martin v. Löwis added the comment:
Ezio: but you would have a green bar "You are logged out - clear this message",
right? And if you follow the "clear this message" link, you get to a URL with
no action, right? So I doubt this is what Terry did.
Terry: can you reproduce t
Martin v. Löwis added the comment:
> > Having a way for non-commiters to indicate to everyone
>
> In XEmacs's Roundup we allow any user to assign themselves an
> "assignable" Role (which is misnamed "Developer"). All this does is
> add them to the
Martin v. Löwis added the comment:
Thanks for the report. This is now implemented in r88945
--
nosy: +loewis
status: unread -> chatting
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/meta/is
Martin v. Löwis added the comment:
This is now fixed in r88935, which updates auth_user when _user changes. In
addition, I manually updated the 187 users which have changed their usernames
since the rietveld setup was created.
Please confirm that it works
Martin v. Löwis added the comment:
I had to delete a lot of rietveld issues due to performance reasons after the
initial setup. I have now also removed the dangling links.
--
nosy: +loewis
status: chatting -> resolved
___
PSF M
Martin v. Löwis added the comment:
This is the tracker for issues with the Python bug tracker. The Python bug
tracker itself is at bugs.python.org.
--
nosy: +loewis
status: unread -> resolved
___
PSF Meta Tracker
&l
Martin v. Löwis added the comment:
This is not the bug tracker for Python, but the bug tracker for Python's bug
tracker, which can be found at bugs.python.org.
--
nosy: +loewis
status: unread -> resolved
___
PSF Meta Tracke
Martin v. Löwis added the comment:
This is not the bug tracker for Python, but the bug tracker for Python's bug
tracker, which can be found at bugs.python.org.
--
nosy: +loewis
status: unread -> resolved
___
PSF Meta Tracke
Martin v. Löwis added the comment:
This is not the bug tracker for Python, but the bug tracker for Python's bug
tracker, which can be found at bugs.python.org.
--
nosy: +loewis
status: unread -> resolved
___
PSF Meta Tracke
Martin v. Löwis added the comment:
This is not the bug tracker for Python, but the bug tracker for Python's bug
tracker, which can be found at bugs.python.org.
--
nosy: +loewis
status: unread -> resolved
___
PSF Meta Tracke
Martin v. Löwis added the comment:
dmalcolm: can you please report whether you received the email from Rietveld,
titled "Rietveld test issue (issue14477)" and posted a minute ago?
___
PSF Meta Tracker
<http://psf.upfront
Martin v. Löwis added the comment:
That's not actually true: dmalcolm *is* in the CC list of
http://bugs.python.org/review/13703/show (at PyCon, we checked the Reviewer's
list, which is a separate list of users in Rietveld that (I think) are the ones
who made reviews).
I created i
Martin v. Löwis added the comment:
I can't reproduce this. It works fine for me.
--
nosy: +loewis
status: unread -> chatting
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/me
Martin v. Löwis added the comment:
I have forwarded this report to postmaster.
--
nosy: +loewis
status: chatting -> resolved
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/meta/is
Martin v. Löwis added the comment:
Retiring the message had not the desired effect; it merely makes it unavailable
for listing. I have now restored the Spam button, and marked the files and the
message as spam.
--
status: chatting -> resol
Martin v. Löwis added the comment:
Closing this report as invalid: the misgeneration of the patch has not been
demonstrated.
--
status: chatting -> resolved
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundu
Martin v. Löwis added the comment:
I suggest to check the specific index, using a custom Python script, to find
out what exactly is in the index.
--
status: unread -> chatting
___
PSF Meta Tracker
<http://psf.upfronthosting
Martin v. Löwis added the comment:
I don't think this is the original issue anymore ("I tried a few keywords that
I know appear in issues, and get no results"), so I think it would have been
best to open a new issue.
As this this issue is now reopened, I degrade it to "b
Martin v. Löwis added the comment:
I have now run pg_upgradecluster. At first it would fail because it ran out of
disk space. I deleted old log files and backups and seefeld's copy of roundup,
so postgres upgrading eventually succeeded.
It turned out that I also had to reindex all ins
Martin v. Löwis added the comment:
What happened is this: before the Debian upgrade, we were using indexer_dbm.
The upgrade installed the Python xapian module, which made indexer_xapian
import correctly, making roundup switch to xapian. There still was a xapian
index around for the tracker
Martin v. Löwis added the comment:
I think I know what happened. I need to take the tracker offline again for
reindexing.
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/meta/issue
Martin v. Löwis added the comment:
Can you provide a specific example? What did you search for, and what should
the result have included?
--
nosy: +loewis
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/meta/issue
Martin v. Löwis added the comment:
Re: init scripts. I added LSB headers to /etc/init.d/roundup
(from the Debian package), and disabled spambayes altogether.
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/meta/issue
Martin v. Löwis added the comment:
> Upgrading the actual virtual host is next.
I notified python-dev about the upcoming outage.
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/meta/is
Martin v. Löwis added the comment:
> Upgrading the actual virtual host is next.
I should be available most of the day tomorrow. So if you can fix a UTC
time, I can try to be available. Else you proceed at your own pace, and
we'll clean up af
New submission from Martin v. Löwis :
We would like to update the psf machine to Squeeze. What is the best procedure
for that?
--
messages: 2394
nosy: izak, loewis, roche
priority: bug
status: unread
title: Upgrade Debian
___
PSF
Martin v. Löwis added the comment:
tshepang: as the tracker operator, I can assert you that I won't do anything
about this for the next year, and likely not for the next five years. If you
really want to see this resolved, submit a patch to the roundup project.
--
nosy: +l
Martin v. Löwis added the comment:
> “utterly broken” is a very strong opinion. The git diff format
> supports things that unified diff does not and has one drawback to my
> knowledge (I forgot exactly what—it was one missing piece of
> information that you mentioned during one o
Martin v. Löwis added the comment:
Thanks!
--
status: chatting -> resolved
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/meta/is
Martin v. Löwis added the comment:
> The current latest changeset on that branch is this merge from default:
> https://bitbucket.org/ncoghlan/cpython_sandbox/changeset/5ce60675e572
The repository you specified on issue 11816 is
http://hg.python.org/sandbox/ncoghlan, not
https://bitbuck
Martin v. Löwis added the comment:
AFAICT, 9512712044a6 *is* the most recent change on the get_opinfo branch.
Which changeset do you think should it have used instead?
___
PSF Meta Tracker
<http://psf.upfronthosting.co.za/roundup/m
Martin v. Löwis added the comment:
So what do you propose should roundup do here? I propose that the current
behavior is actually "correct", and that mismatches with user expectations
should be dealt with by education.
For the patch generation to work in the expected manner, you co
Martin v. Löwis added the comment:
The diff generator will always generate a patch against some revision of
cpython. IIUC, you would want it to diff against a base revision that is *not*
in cpython.
I'm doubtful that this is a desirable feature:
a) your patch cannot be applied as-
Martin v. Löwis added the comment:
I think some indication must be given to a legitimate user, as the user
otherwise may not recall what email account to check. In the specific case of
bugs.python.org, it may, in particular, be a sourceforge address.
If people are worried that users
Martin v. Löwis added the comment:
Using the mercurial diff format is intentional - the git format is utterly
broken and must not be used.
As for changes that are reverted: what changes specifically? (or: what versions
should it have compared instead of comparing 0c1c9bb590a9 and
New submission from Martin v. Löwis :
http://psf.upfronthosting.co.za is currently off by ca 15s from UTC. This broke
my new OpenID login procedure, which assume a maximum offset of 10s. I worked
around by allowing up to 5 minutes.
Can time synchronization be improved?
--
assignedto
1 - 100 of 164 matches
Mail list logo
