On Mon, Apr 20, 2020 at 9:36 AM Johan Corveleyn wrote:
> Let's start a new thread with the bits of information I posted about
> building on Windows in the "Release Notes: TODO regarding Py3 support"
> thread [1].
>
> Building SVN on Windows is pretty hard, especially the dependencies. A lot
> of
On Wed, Sep 16, 2015 at 8:44 AM, Ivan Zhakov wrote:
> Just to keep everyone updated: ASF Infra team performed first test
> migration for us:
> https://issues.apache.org/jira/browse/SVN
>
> Please report me any problems or suggestions. Current TODO list:
> - Migrate attachments
> - Preserve history
What's the status code from serf_bucket_read() on line 1151? I would
expect that the socket read should be generating a ECONNABORTED
somewhere? -- justin
On Wed, Nov 5, 2014 at 9:24 AM, Philip Martin wrote:
> Configure a server with both mod_ssl (HTTPS) and mod_deflate (HTTP
> compression).
On Wed, Sep 10, 2014 at 1:06 PM, Stefan Sperling wrote:
> I've been working on the log-message-templates branch recently.
> This branch was started by cmpilato a long time ago but for some
> reason sat around untouched ever since. It's now been overhauled
> by me with much help from Bert.
>
> I be
On Thu, Aug 21, 2014 at 5:30 PM, Lieven Govaerts wrote:
> Attached patch implements a slightly different approach but I think
> it's a valid workaround for the problem.
It may be useful to note why we register apps_ssl_info_callback in the
re-negotiate info callback. (It's because there can only
On Tue, Aug 19, 2014 at 2:53 PM, Lieven Govaerts wrote:
> However, when the client certificate is requested for a resource
> deeper in the repository, it's likely that say during a large
> checkout, many (pipelined) requests will already be sent by the client
> before the request for the protected
On Mon, Jul 21, 2014 at 3:26 AM, Stefan Fuhrmann
wrote:
> So, it turned out to be a problem in the Subversion libs and
> got fixed by r1611379. When multiple connections to the same
> repo were opened concurrently (ra_serf over loopback), our file
> API workarounds for Windows (retry for up to 10-
On Mon, Jul 14, 2014 at 9:09 AM, Stefan Fuhrmann
wrote:
> On the same machine actually (which may be a contributing factor).
> The client is svn-bench that simply handles the editor drive but
> discards incoming file contents etc.
We've historically seen oddity in loopback scenarios on Windows -
On Sun, Jul 13, 2014 at 7:54 PM, Stefan Fuhrmann
wrote:
> * ra_serf has an anomaly making "fast" config up to 100x slower
> than ra_svn - independent of repo format
...
> After that, I'll try to isolate the trigger for the ra_serf anomaly.
Huh. Very curious to read more about this when you get
On Mon, Apr 7, 2014 at 12:03 PM, Ben Reser wrote:
> So to summarize. 3 votes must come from separate individuals. Our platform
> testing requirements do not have to be met by separate people (though each
> person obviously can only have one vote for each platform).
+1.
To be pedantic, your las
On Tue, Apr 1, 2014 at 3:17 PM, Ben Reser wrote:
> It's proving to be impossible to move forward with the alpha2 release.
>
> Let's officially say that alpha and beta releases only require 3 votes
> (platform doesn't matter). This does not apply to release candidates.
+1. (I've always maintaine
On Mon, Mar 24, 2014 at 1:35 PM, Julian Foad wrote:
> But, compared with the present 1 ms that's inadequate on many systems and the
> 1 second that's too much on some but not enough on others, I would be happy
> with simply increasing the 1 ms to 10 ms for a "quick fix" if we don't want
> to in
On Tue, Feb 4, 2014 at 6:12 AM, Branko Čibej wrote:
> I've changed my mind since and now firmly stand by the position that I
> never, ever suggested anything other than having a no-pager default. :)
>
> I'm also somewhat opposed to the idea of having pager configuration crud for
> each client subc
On Sun, Nov 17, 2013 at 7:34 AM, Bert Huijben wrote:
> But I would also like to recommend/ask that we start bundling Sqlite with
> Subversion, to allow optimizing for the specific version we use for a
> release without risking future breakage.
>
If we use the amalgamation, are all of the symbols
On Wed, Aug 7, 2013 at 5:48 AM, Daniel Shahaf wrote:
> (v3)
> - Receive report
> - Come up with a fix
> - Gather 3 +1 votes via shadow status file
> - Send pre-notifications
> - Release a signed .diff file (instead of a tarball) as 1.8.(x+1)
> - Commit fix with a log message clearly outlining the
On Sat, Jul 13, 2013 at 8:39 PM, Daniel Shahaf wrote:
> It appears as soon as svn_utf_cstring_to_utf8() is called --- which is
> normally during argv parsing. The error happens even if the argv
> arguments are all ASCII, which effectively adds a new dependency on
> apr_xlate_* even for people wh
On Thu, Jul 11, 2013 at 10:15 PM, Greg Stein wrote:
> Hey all,
>
> Right now, trunk and the 1.8.1 backport request implements the following
> option:
>
> http-detect-chunking = off / default: perform no detection. A 411
> response will throw an error message suggest to turn this option on.
>
I'm
On Mon, Jul 1, 2013 at 5:37 PM, Daniel Shahaf wrote:
> How about adding 'svn cleanup --rm-I' as a short option for 'svn cleanup
> --remove-ignored'?
>
I don't like either option (rm-I or rm-?) as I don't find either an
improvement over the long option. Mixing lower-and-upper in the same
argument
+1 for "no big deal" too, I think.
Depending upon how far we take serf 2.0 and its APIs, a new libsvn_ra_serf2
library may be in order. So, it'd be no different than all of the
libsvn_fs* and libsvn_wc internal changes.
If we altered the RA APIs, then that would be a no-no if we did it in a
back
On Wed, May 29, 2013 at 6:03 PM, Blair Zajac wrote:
> Yup, I've had lots of issues with this. Putting C++ pool wrappers in C++
> classes and having them destroy in the correct order can be tricky to get
> right (lots of core dumps in our internal RPC server). One of the nice
> things about movi
On Wed, May 29, 2013 at 12:05 PM, Blair Zajac wrote:
> I'm generally in favor of a move to C++, it would be nice to get features
> that we work around now in C.
>
Rewriting even some of our core libraries to use C++ (even if it we kept
the existing C API) just doesn't seem to address any real pr
On Fri, May 17, 2013 at 5:58 PM, Ivan Zhakov wrote:
> Yes, this will be good improvement anyway, but I think repository
> integrity should be first goal.
>
+1! =P -- justin
On Wed, Apr 24, 2013 at 1:34 AM, Branko Čibej wrote:
> If you consider all this, the easiest approach by far might be to simply
> add a Lucene index of all log messages to the server, then you can and
> any number of bells and whistles including language-specific stemming.
> I'd consider that a
On Sun, Apr 21, 2013 at 9:37 AM, Bert Huijben wrote:
> Do we want case folding (or at least case variant compare) support in our
> libraries for 1.8?
>
No.
> Or is this 1.9+ scope?
>
Yes. =)
My $.02. -- justin
On Fri, Apr 12, 2013 at 6:28 PM, Peter Samuelson wrote:
> What is the utility of the "(apr_err=_)" suffix there? I think it's
> better to omit it. Print apr_err= only when it is new information.
>
+1. -- justin
On Tue, Feb 26, 2013 at 1:54 PM, C. Michael Pilato wrote:
> * The appropriate time to stop supporting Berkeley DB is in the same
> release
> for which existing FSFS will also have to dump/load. It is cruel to force
> admins to endure the migration process twice -- possibly in successive
> release
On Feb 18, 2013 7:08 PM, "Stefan Sperling" wrote:
>
> On Mon, Feb 18, 2013 at 10:21:16PM +0100, Bert Huijben wrote:
> > If I would use 1.8 to run
> >
> > svn diff |more
> >
> > And leave that open in a console I would block TortoiseSVN, AnkhSVN,
Subclipse, etc on my entire working copy, while this
On Sun, Feb 3, 2013 at 6:42 AM, Stefan Sperling wrote:
> On Sun, Feb 03, 2013 at 06:16:03AM -0500, Justin Erenkrantz wrote:
> > On Sun, Feb 3, 2013 at 6:06 AM, Justin Erenkrantz >wrote:
> >
> > > I thought we treated application/xml as text at one time? -- justi
On Sun, Feb 3, 2013 at 6:06 AM, Justin Erenkrantz wrote:
> I thought we treated application/xml as text at one time? -- justin
>
Yes, we did...and Ben -1'ed it back in 2004. See:
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=374218
I guess...but...
On Sun, Feb 3, 2013 at 4:41 AM, Bert Huijben wrote:
> Maybe we should try to fix this automatic detection then (Does your
> setting match the libmagic output for that file?) instead of adding
> just a warning when setting the file binary explicitly.
>
> Maybe we should validate our libmagic based
As another data point, I have hit this text-as-binary myself just a few
weeks ago when I added a bunch of HTML files to a local repository - so,
it's definitely occurring automatically. I did not have a chance to dig
into why the magic detection failed so miserably... -- justin
On Saturday, Febr
FWIW, I'll be at FOSDEM in Brussels this weekend (arriving on Friday
morning) - if any other SVN devs are going to be there, it'd be fun to
catch up.
Cheers. -- justin
On Tue, Jan 29, 2013 at 1:04 PM, Ben Reser wrote:
> If we're not going to keep these things up to date then I think we
> should just remove them.
>
+1 to remove. (It's all version-controlled anyway!) -- justin
On Fri, Jan 18, 2013 at 7:54 AM, Branko Čibej wrote:
> Doesn't help ... apparently because we get both APR_POLLIN and
> APR_POLLHUP events, and the former is processed first, the function returns
> without error in the previous block:
>
> /* If we decided to reset our connection, return
On Thu, Jan 17, 2013 at 1:19 AM, Branko Čibej wrote:
> > 1 has not been marked as started nor been assigned but is tied to an
> > open serf issue:
> > 4274 DEFECT P3 All issues@subversion philip NEW
> 1.8.0 serf
> > client hangs when server crashes
>
> I reviewed this th
On Wed, Jan 9, 2013 at 3:30 AM, Branko Čibej wrote:
> We talked about that a couple days ago, but the problem is that a
> Windows VM requires a Windows OS license for every user of that VM.
> That's not something we can provide or hack around (well ... we could
> hack around it, but it would be a
On Tue, Jan 8, 2013 at 10:07 PM, Ben Reser wrote:
> Given stefan2's argument I don't think it's unreasonable to lower the
> Windows votes, but I think removing them entirely is probably going
> too far.
>
Given the fact that we repeatedly have trouble securing votes just for
Windows, I think you
IMO, we should follow most other (all?) ASF projects - you need 3 +1s for
release regardless of platform. The fact that we require 3 +1s just for
Windows is very odd - we don't require 3 +1s for Mac and 3 +1s for RHEL and
3 +1s for Ubuntu, etc. -- justin
On Jan 8, 2013 3:29 PM, "Ben Reser" wrote
On Sun, Jan 6, 2013 at 9:44 PM, C. Michael Pilato wrote:
> hosted elsewhere for them. The BDB backend (thanks to improvements to the
> Berkeley DB library itself) is much more stable today than it was when we
> first started this project, so it's quite possible that we don't hear noise
>
That's
+1! -- justin
On Jan 5, 2013 10:40 AM, "Greg Stein" wrote:
> Is "+1" too short of a response?
>
> :-)
> On Jan 4, 2013 7:35 PM, "Branko Čibej" wrote:
>
>> For the following reasons
>>
>>- FSFS has been the default filesystem backend for almost 7 years,
>>since 1.2.
>>
>> - Looking
ustin
On Jan 3, 2013 2:18 AM, "Peter Samuelson" wrote:
>
> [Justin Erenkrantz]
> > Because GNU libtool always compiles code twice on Mac OS X (insert rant
> > against GNU libtool here)
> ...
> > But, then again, I always like reminding devs how evil lib
I'm not entirely sure if we should commit this, but I'll throw this on-list
for posterity and future reference.
I was trying to compile trunk and was getting the following:
/bin/sh path-to-svn/libtool --tag=CC --silent --mode=compile ccache gcc
-std=c89 -DDARWIN -DSIGPROCMASK_SETS_THREAD_MASK -n
On Tue, Jan 1, 2013 at 9:43 PM, Miriam Hochwald wrote:
> Hello Justin,
>
> Actually the INSTALL docs ... whilst detailed make me phobic. It seems to
> target the audience that likes to tinker, rather than just click and oh ...
> there it is (think iPad users ... which is most of the world). It mi
On Mon, Dec 31, 2012 at 4:22 AM, Miriam Hochwald
wrote:
> I am new to this random public email writing ... and, at least for me it
> takes a fair amount of courage to do so. I have had to face a few phobia
> barriers based on experience.
>
Welcome! It can indeed be quite scary at first...then, y
On Tue, Jan 1, 2013 at 4:17 PM, Gabriela Gibson
wrote:
> The following is a conversation that should have been held on the
> list, please excuse the misunderstanding on my part.
>
No worries - we've all been there before. You'll get the hang of it soon
enough. =) Keep up the posts!
Cheers. -
On Mon, Dec 31, 2012 at 9:05 AM, Stefan Sperling wrote:
> BTW, it would have been better to send your question to the users@ list.
> The dev@ list is off-topic for questions like this. I suppose you chose
> dev@ because of your idea to develop support for cross-repository copies,
> which I'll add
On Wed, Dec 12, 2012 at 5:06 PM, Daniel Shahaf wrote:
> P.S. This thread was an unusually long one, for a patch that adds about
> a dozen lines of code.
>
Uh, how is that at all unusual for this crowd? =) -- justin
On Wed, Dec 12, 2012 at 4:24 PM, C. Michael Pilato wrote:
> Here's a question I've been wondering for some time: should we expose to
> users a configuration option for declaring the number of aux connections
> ra_serf should use? I mean, Firefox has exposed such an option for years.
> If we did
On Wed, Dec 5, 2012 at 4:50 PM, Stefan Sperling wrote:
> In the 1.8 release notes We can raise attention in the fact that a 1.8
> server should turn off skelta mode if necessary, such as when each GET
> request causes an authentication request to LDAP.
>
+1. I'm all for documenting and giving k
On Sun, Dec 2, 2012 at 9:05 AM, Lieven Govaerts wrote:
> Attached the patch. ( I get paid per mail I send to this list in case
> no one noticed. ;) )
The patch looks right to me - the short-circuit to do the simple window can
clearly be called for any window regardless of offset. Perhaps we ca
On Sat, Dec 1, 2012 at 12:00 PM, Mark Phippard wrote:
> I feel pretty strongly that we should at minimum use the send-all
> approach when talking to pre-1.8 servers. Even though in some
> situations it could still offer good performance. I just think it
> would be more respectful to our users (
On Sat, Dec 1, 2012 at 9:41 AM, Branko Čibej wrote:
> On 01.12.2012 14:31, Justin Erenkrantz wrote:
> > And, yes, that clearly could all be done in time for 1.8 without
> > jeopardizing the timelines one tiny bit. =P
>
> Eep ... :)
>
>
> Another thing I've been
On Sat, Dec 1, 2012 at 9:28 AM, Daniel Shahaf wrote:
> BTW, the ability to change svn:author at will is one of the reasons they
> aren't global-scoped: if Subversion ever migrated away from ASF, we can
> _then_ change all svn:author revprops --- just like we once changed
> "zhakov" (implied @tigri
On Sat, Dec 1, 2012 at 9:01 AM, Lieven Govaerts wrote:
> There are some scenario's where either the server admin or the user
> can decide if parallel requests make sense or not.
>
> I'm specifically thinking of the use Kerberos per request
> authentication. These responses can't be cached on the
On Sat, Dec 1, 2012 at 8:53 AM, Eric S. Raymond wrote:
> I'm not certain, but if your server-side hooks work the way I think
> they do, all of this except (1) can be done in Python. Not having to
> add complexity to your C code is a significant virtue.
>
Here's another approach to take with reg
On Sat, Dec 1, 2012 at 5:59 AM, Johan Corveleyn wrote:
> I'm wondering whether your concerns apply to both internet-wide
> deployments and local (all on the same LAN) ones.
>
That line is certainly a fair one to draw in the sand. That said, I think
the internal use case cries out even *more* fo
On Fri, Nov 30, 2012 at 5:26 PM, Branko Čibej wrote:
> When I create an account at [your favourite forge], I tell it my name
> and give it one of my e-mail addresses. It is, in my opinion, up to the
> forge software to use that in svn:author, not up to local user
> preferences. That /is/ a fundam
On Fri, Nov 30, 2012 at 4:54 PM, wrote:
> Author: cmpilato
> Date: Fri Nov 30 21:54:35 2012
> New Revision: 1415864
>
> URL: http://svn.apache.org/viewvc?rev=1415864&view=rev
> Log:
> Implement in ra_serf "send-all" mode support for update-style REPORTs
> and their responses. (Currently disabled
On Tue, Nov 13, 2012 at 2:26 AM, Lieven Govaerts wrote:
> Is that a fix that we can get in apache 2.2? Justin?
>
I don't readily recall the issue in mod_deflate. Is it already resolved in
the 2.4 series?
> work. These concern maybe only 1% of the users, but we should have a
> plan to support
On Tue, Nov 13, 2012 at 2:41 AM, Lieven Govaerts wrote:
> The proposed approach here is to disable SSL compression completely,
> so in terms of the above that leaves us with scenario 2.
>
> Serf doesn't have an option currently to disable SSL compression from
> the client side. I plan to add it i
On Mon, Nov 12, 2012 at 9:13 PM, Philip Martin
wrote:
> I have a checkout of the gcc tree, it has 78,000 files. Now it uses
> svn: but if it were to use http: then the serf checkout log would be 4
> orders of magnitude bigger than the neon log. 83 years becomes 1 or 2
> days.
>
> The neon log is
On Sat, Nov 10, 2012 at 3:25 PM, Thomas Åkesson wrote:
> I suppose this means that it would be a significant optimization to
> perform HEAD rather than GET when discovering ACLs for every subdirectory
> in a directory listing?
>
Probably - doing the HEAD request will run the full authn and authz
On Sat, Nov 10, 2012 at 6:49 AM, Justin Erenkrantz wrote:
> There is a C-L header...so, I don't know what the original poster is
> seeing, but we're already doing the right thing... -- justin
>
I bet the OP is trying to do HEAD on a directory as there was talk
elsethr
On Fri, Nov 9, 2012 at 11:37 PM, Branko Čibej wrote:
> That would imply that, if content-length doesn't get set on a HEAD
> response, but Transfer-Encoding: chunked does, then everything is
> correct, right? If somewhat inefficient.
>
The chunked header is only sent on non-HEAD responses. r->ch
On Fri, Nov 9, 2012 at 10:15 AM, C. Michael Pilato wrote:
> request, so I'm wondering ... does Apache just handle a HEAD as a GET
> under-the-hood and then discard the resulting response body? The comment
>
Unless the handler does something special, yes, httpd will treat HEAD as a
GET until it i
On Tue, Nov 6, 2012 at 12:13 PM, Lieven Govaerts wrote:
> okay, so with the Timeout directive added I can reproduce this issue.
>
> What I see is that the server closes the connection in the middle of
> sending a response to the client. It doesn't even finalize the
> response first.
> So ra_serf i
On Tue, Nov 6, 2012 at 4:41 PM, Ivan Zhakov wrote:
> Another problem is how serf reads data from network in case of
> multiple connections: it reads data from one connection until EAGAIN.
> But if data comes from network really fast (from local server for
> example) it continue reading data from t
On Tue, Nov 6, 2012 at 3:22 PM, C. Michael Pilato wrote:
> On 11/06/2012 10:15 AM, C. Michael Pilato wrote:
>> On 11/06/2012 08:29 AM, C. Michael Pilato wrote:
>>> I recall this patch of yours -- I even asked you about it post-Berlin. I'll
>>> take a look at it now and see if it can help us out.
On Mon, Nov 5, 2012 at 1:33 PM, C. Michael Pilato wrote:
> My debugging indicates that when close_all_dirs() is called, there are a
> slew of unclosed directories which each have a non-zero ref_count. Save for
> the root directory of the edit (/tags), fetch_props == TRUE,
> propfind_handler->done
On Sat, Oct 13, 2012 at 6:30 PM, Ben Reser wrote:
> On 10.7:
> XFAIL: op-depth-test 23: test_case_rename on case (in)sensitive system
> XFAIL: copy_tests.py 102: case-only rename
>
> with:
> $ apr-1-config --version
> 1.4.2
As a follow-up for those not present here in CT, Ben wasn't exactly on
th
On Sat, Oct 13, 2012 at 5:35 PM, Bert Huijben wrote:
> Which file system do you run your tests on? (it is easy to create case
> sensitive filesystems manually, but there is no platform independent
> way to detect them)
It's hfs+ case-insensitive - and I am pretty sure that these tests
didn't XPAS
When I run make check on Mac OS X 10.8.2, it looks like the
case-sensitivity checks now XPASS:
XPASS: op-depth-test 23: test_case_rename on case (in)sensitive system
XPASS: copy_tests.py 102: case-only rename
I'm wondering if the case-preserving semantics in 10.8 changed - as it
looks like rename
On Tue, Sep 18, 2012 at 7:24 AM, Philip Martin
wrote:
> That sounds like the wrong behaviour as well. By default we should
> build as many stable features as possible. If I pass an explict option
> to enable/disable a feature that should be a hard requirement.
+1. -- justin
On Thu, Aug 30, 2012 at 7:22 PM, Daniel Shahaf wrote:
> Which _shell_ you were using had an impact? I can see how the terminal
> emulator in use, the $TERM value, etc would have an impact... but why
> does the identity of svn's parent process have any effect?
Yes, the underlying terminal emulato
On Thu, Aug 30, 2012 at 3:11 AM, Johan Corveleyn wrote:
> But more to the point: anybody have a solution in mind? If it's not
> Windows-only then some Windows defines wont help of course. Buffering
> the output may be the only way to eliminate this bottleneck? What are
> the pros and cons, and how
On Wed, Aug 29, 2012 at 4:04 PM, C. Michael Pilato wrote:
> I misremembered Greg and Justin's attitude toward my approach, thinking they
> were just flatly opposed. As I re-read the relevant threads, though, I
> think it's clear that perhaps both my approach and their PROPFIND-Depth-1
> approach
On Tue, Aug 28, 2012 at 7:22 PM, Johan Corveleyn wrote:
> Yep, redirecting to a file eliminates the bottleneck (almost the same
> as redirecting to NUL) (I ran it a couple of times to make sure the
> server cache was hot):
FWIW, I've historically seen similar behavior on Unix platforms as
well -
On Fri, Aug 24, 2012 at 9:56 AM, C. Michael Pilato wrote:
> If it makes you feel better, I've been down this road before, too. :-)
*grin*
> One thing that occurs to me as possible solution is to add a flag to
> svn_ra_openX() which instructs the RA layer to immediately negotiate the
> current y
On Fri, Aug 10, 2012 at 2:20 AM, C. Michael Pilato wrote:
> Justin, you appear to have resurfaced (to a degree, at least). Did you ever
> complete this work?
No.
AFAICT, the rewrite of the XML parsing used in update.c and the Depth:
1 PROPFIND hasn't been completed either. I think it probably
On Wed, Aug 8, 2012 at 1:40 PM, Philip Martin
wrote:
> Subversion 1.7.6 tarballs are now available for testing/signing by
> committers. To obtain them please check out a working copy from
> https://dist.apache.org/repos/dist/dev/subversion
+1 for release.
Tested on Mac OS X 10.7.4.
All tests pa
On Tue, Aug 7, 2012 at 5:23 AM, Philip Martin
wrote:
> This thread has gone quiet. I plan to roll a tarball tomorrow,
> Wednesday, for release next week.
I wouldn't have gotten to it until Thursday at the earliest. Thanks. -- justin
On Thu, Aug 2, 2012 at 6:10 PM, Hyrum K Wright wrote:
> So, I'm still on semi-hiatus, but I've been seeing all this stuff
> being merged to 1.7.x and I can't help but wonder what the plan for
> the 1.7.6 release is. I haven't found any discussion on list. The
> release process has changed enough
One issue that I've come across in my recent analysis is that ra_serf
isn't always aggressive in closing directories when doing an update.
Specifically, when the directory propfind isn't received by the time
all of the files are processed, ra_serf won't look at the directory
handle again until ever
On Thu, Jun 14, 2012 at 5:02 PM, Daniel Shahaf wrote:
> I know pf(4) on openbsd has some specific "Be unreliable" options.
> Would they be useful for this?
Since we're using TCP, I expect that such pf filters would just cause
delays that TCP would essentially recover from and might not even be
vi
On Thu, Jun 14, 2012 at 2:06 PM, Justin Erenkrantz
wrote:
> I checked the wireshark traffic on a checkout and it looks like all
> the right headers are set - with this patch, we now have
> Cache-Control, ETag, and Last-Modified all set. When mod_deflate is
> run, it'll set the
On Thu, Jun 14, 2012 at 3:14 PM, Johan Corveleyn wrote:
> Great! I can now successfully run the entire testsuite over ra_serf
> with AVG SurfShield enabled. For the first time ever :-).
Yay!
> Thanks for all the bugfixes.
No problem. Thanks for providing the underpowered laptop with the
nutty
Hey, lookie, a bunch of ###'s. =)
I'm not quite sure whether we would want to set Cache-Control to just
"public" and not specify a max-age. Or, perhaps we need to add some
configuration directives to allow the admin to tweak the value.
I checked the wireshark traffic on a checkout and it looks
On Wed, Jun 13, 2012 at 11:28 PM, Johan Corveleyn wrote:
> With latest trunk (@1349944) I get a crash in authz_tests.py 15
> ("authz issue #3242 - access required at repo root") when running with
I have confirmed on your machine that the fix in r1350179 doesn't
cause a crash any more.
Thanks! -
handler->handler_pool is somehow NULL when we invoke get_locations. Weird.
Perhaps you can throw some printf's before the get_locations call to see
if it is indeed NULL and not an artifact of the trace. A quick glance at
the code shows handler_pool should be set in add_file correctly. -- justin
(I do not have an easy way to login to the issue tracker from my iPad...so,
moving back to the list...)
I really do not think the assert is due to memory corruption - nothing I
saw indicated that was happening. I think it is far more likely that there
is another bug somewhere. (It is the simples
On Wed, Jun 13, 2012 at 2:14 AM, Johan Corveleyn wrote:
> So maybe we can take another look at this tomorrow (unless someone
> else fixes it first of course :-)).
It happens here on my Mac too - should be fixed in r1349651. Thanks! -- justin
Lock test #37 and #42 is currently failing on Windows (with and
without the virus scanner). See:
http://ci.apache.org/builders/svn-slik-w2k3-x64-ra/builds/4511/steps/Test%20fsfs+serf/logs/faillog
It is related to the special 405/409/500 handler in handle_response
which is "completely wrong". *g
On Tue, Jun 12, 2012 at 2:38 PM, Lieven Govaerts wrote:
> Attached patch shows what I suggest. The case you probably encounter
> is when the response is handled by handle_server_error.
> This is untested, I don't have a Windows setup ready and didn't
> install my build tools yet after upgrade to O
On Tue, Jun 12, 2012 at 1:56 PM, Lieven Govaerts wrote:
>> How is it supposed to retry? ra_serf needs to let the context_run
>> loop execute again to pull off the remaining bits from the socket.
>> Without the patch, we treat EAGAIN as a fatal error and it gets
>> returned all the way to the clie
On Tue, Jun 12, 2012 at 1:50 PM, Johan Corveleyn wrote:
> Justin, here's the patch for the 404 which you made on my machine ...
I'm working on cleaning up this patch...but, with this patch plus the
patch at the beginning of the thread, all of the basic tests now pass
on Johan's machine with the f
On Tue, Jun 12, 2012 at 1:01 PM, Lieven Govaerts wrote:
> That seems strange, serf_context_run is not documented to return
> APR_EGAIN. While the response handler can return APR_EAGAIN, serf
> itself will translate this to APR_SUCCESS. As a result, ra_serf
> doesn't expect it and - rightfully IMO
It seems we've tweaked how we handle EAGAIN inside of ra_serf and are
treating it as a critical error.
Right now, what Johan is seeing with his anti-virus software is that
we very often receive EAGAIN from the sockets. However, we're
treating EAGAIN as a fatal error inside of
svn_ra_serf__context
On Tuesday, June 12, 2012, C. Michael Pilato wrote:
> On 06/10/2012 12:16 AM, Bert Huijben wrote:
> >> @@ -1214,7 +1215,11 @@ start_xml(void *userData, const char *ra
> >>
> >>svn_ra_serf__expand_ns(&name, parser->state->ns_list, raw_name);
> >>
> >> - parser->error = parser->start(parser, na
On Tuesday, June 12, 2012, C. Michael Pilato wrote:
> On 06/10/2012 09:13 AM, Justin Erenkrantz wrote:
> > One question: what happens when the WC's pristine gets corrupted. That
> is,
> > we originally get the pristine properly, stash it in the local DB, but
> > som
x27;t
process properties until we open the file through the editor drive much
later on. So, we may give back some of the performance that way when we
get all of the WC props up front. -- justin
On Jun 11, 2012 10:58 AM, "Ivan Zhakov" wrote:
> On Sun, Jun 10, 2012 at 9:19 AM, Justin E
1 - 100 of 295 matches
Mail list logo
