Issues - drogus/apache-upload-progress-module

drogus / apache-upload-progress-module

117

Description:

Upload progress module for apache

Sort by: Priority | Votes | Last Updated

Loading…

1.

0 votes

Apache/Passenger pair returns "Internet Server Error" when uploading large files
0 comments Created 4 months ago by ekontsevoy

Title

Body
When I upload large (5-25MB) files, I get "Internal Server Error" at the end of download. Here is the apache's error.log: http://pastie.org/522860 It looks like the problem lies in mod_passenger, but when I disable upload_progress_module, everything starts working as usual. As you can see from the log, I am running Apache/2.2.8 on Ubuntu 8.04 (latest LTS) with the latest mod_passenger and Ruby EE. Playing with UploadProgressSharedMemorySize parameter helps a little, but I can always find a file which won't upload. Feel free to contact me at ekontsevoy *at* profista *dot* com if you need help reproducing/fixing the issue. Regards, Eugueny Kontsevoy EDIT: nevermind, I have upgraded to the latest Ruby EE and the issue went away. I can't find how can I delete it from Github though.
or cancel

When I upload large (5-25MB) files, I get "Internal Server Error" at the end of download.
Here is the apache's error.log: http://pastie.org/522860

It looks like the problem lies in mod_passenger, but when I disable upload_progress_module, everything starts working as usual.

As you can see from the log, I am running Apache/2.2.8 on Ubuntu 8.04 (latest LTS) with the latest mod_passenger and Ruby EE.

Playing with UploadProgressSharedMemorySize parameter helps a little, but I can always find a file which won't upload.

Feel free to contact me at ekontsevoy at profista dot com if you need help reproducing/fixing the issue.

Regards,
Eugueny Kontsevoy

EDIT: nevermind, I have upgraded to the latest Ruby EE and the issue went away. I can't find how can I delete it from Github though.

Comments

Parsed with GitHub Flavored Markdown
2.

1 vote

Segmentation fault (11) when mod_upload_progress loaded
3 comments Created 3 months ago by jonn

Title

Body
When using the latest apache upload progress module (with changes as recent as 5-21-09) about 30% of uploads fail with a message like the following in the apache error_log: <pre> [Wed Jul 15 15:45:23 2009] [notice] child pid 29196 exit signal Segmentation fault (11) </pre> Requesting the progress after this happens with the corresponding X-Progress-ID continually returns a status of "starting". I am running: Apache 2.2.11 w/ Passenger 2.2.4/Rails 2.3.2 I'm also using mod_rewrite and mod_xsendfile along with the default enabled modules of Apache. I tried recompiling with --enable-exception-hook and loading mod_whatkilledus to get more info (per http://www.techds.in/2009/02/debugging-notorious-segmentation-fault.html) but was unable to gather any more useful information. Recompiled again without --enable-exception-hook but kept the --enable-nonportable-atomics=yes option and the upload failures continue, but the corresponding error log entries now look like this: <pre> *** glibc detected *** double free or corruption (out): 0x086f9b08 *** [Fri Jul 17 10:25:46 2009] [notice] child pid 6633 exit signal Aborted (6) </pre> When I disable mod_upload_progress, uploading works consistently again. Sorry I can not provide more helpful information. - Jonn
or cancel
When using the latest apache upload progress module (with changes as recent as 5-21-09) about 30% of uploads fail with a message like the following in the apache error_log:
```
[Wed Jul 15 15:45:23 2009] [notice] child pid 29196 exit signal Segmentation fault (11)
```
Requesting the progress after this happens with the corresponding X-Progress-ID continually returns a status of "starting".

I am running:
Apache 2.2.11
w/ Passenger 2.2.4/Rails 2.3.2

I'm also using mod_rewrite and mod_xsendfile along with the default enabled modules of Apache.

I tried recompiling with --enable-exception-hook and loading mod_whatkilledus to get more info (per http://www.techds.in/2009/02/debugging-notorious-segmentation-fault.html) but was unable to gather any more useful information.

Recompiled again without --enable-exception-hook but kept the --enable-nonportable-atomics=yes option and the upload failures continue, but the corresponding error log entries now look like this:
```
*** glibc detected *** double free or corruption (out): 0x086f9b08 ***
[Fri Jul 17 10:25:46 2009] [notice] child pid 6633 exit signal Aborted (6)
```
When I disable mod_upload_progress, uploading works consistently again.

Sorry I can not provide more helpful information.
- Jonn
Comments
jonn Tue Jul 21 12:27:12 -0700 2009 | link
A little more data...

I thought I'd try adding mod_porter to the mix, thinking maybe the upload progress module would play nicer with mod_porter in-between it and passenger. Occasional failures continue, but now the error logged is:

*** glibc detected *** free(): invalid pointer: 0x09264450 *** [Tue Jul 21 13:20:21 2009] [notice] child pid 2133 exit signal Aborted (6)
Comment
A little more data... I thought I'd try adding mod_porter to the mix, thinking maybe the upload progress module would play nicer with mod_porter in-between it and passenger. Occasional failures continue, but now the error logged is: <pre> *** glibc detected *** free(): invalid pointer: 0x09264450 *** [Tue Jul 21 13:20:21 2009] [notice] child pid 2133 exit signal Aborted (6) </pre>
or cancel
expandrive Thu Sep 10 08:52:43 -0700 2009 | link

I'm seeing a similiar issue, Jonn. Are you running on open solaris?

Comment
I'm seeing a similiar issue, Jonn. Are you running on open solaris?
or cancel
jonn Mon Sep 14 15:25:50 -0700 2009 | link
No, I'm on a WestHost provided Linux VPS instance. I believe it is RedHat derived.

I'm still having the problem but I have not dug any deeper. I watch for the problem on the client and restart the upload when the server process crashes. This has been acceptable for the short-term, but I'll need to chase down the problem eventually. Please let me know if you make any progress on resolving the problem. I'll do the same.

Jonn
Comment
No, I'm on a WestHost provided Linux VPS instance. I believe it is RedHat derived. I'm still having the problem but I have not dug any deeper. I watch for the problem on the client and restart the upload when the server process crashes. This has been acceptable for the short-term, but I'll need to chase down the problem eventually. Please let me know if you make any progress on resolving the problem. I'll do the same. - Jonn
or cancel
Parsed with GitHub Flavored Markdown
3.

0 votes

Doens't work with mod_mem_cache
2 comments Created 2 months ago by isaiah

Title

Body
I think should put this issue in the README file, or fix it.
or cancel

I think should put this issue in the README file, or fix it.

Comments

drogus Wed Sep 02 04:10:18 -0700 2009 | link

Could you give a little bit more details? Does it crash, or not reports uploads, or maybe somehing else?

Comment
Could you give a little bit more details? Does it crash, or not reports uploads, or maybe somehing else?
or cancel

isaiah Wed Sep 02 04:32:54 -0700 2009 | link

I'm using Archlinux and the /progress keeps hang until I found this:

Slobodan said

Jan 12, 2009 @ 12:50 AM

Peter, I debugged it and think I figured out what it was. My Apache configuration (2.2, ArchLinux) was loading all sorts of modules, including mod_mem_cache. Now for some reason the call to cache_free in line 434 was always crashing. To reproduce, I uploaded a file, waited 60s for the node to expire and submitted another file, at which point the worker process crashed at the call to cache_free.

From the backtrace I could see that the call to cache_free was calling the function with the same name, but in the mod_mem_cache, instead of the local one in the mod_upload_progress. How this was possible I still have no idea (or maybe I didn’t understand the code enough). As soon as I disabled mod_mem_cache, the progress tracking was working like a breeze.

It's in the comments of your blog post(http://drogomir.com/blog/2008/6/18/upload-progress-bar-with-mod_passenger-and-apache), and you said you are gonna to fix this issue. As Slobodan advised, I disabled mod_mem_cache, and progress tracking works. I didn't dig into the source code so I have no idea what's wrong.

Comment
I'm using Archlinux and the /progress keeps hang until I found this: Slobodan said Jan 12, 2009 @ 12:50 AM Peter, I debugged it and think I figured out what it was. My Apache configuration (2.2, ArchLinux) was loading all sorts of modules, including mod_mem_cache. Now for some reason the call to cache_free in line 434 was always crashing. To reproduce, I uploaded a file, waited 60s for the node to expire and submitted another file, at which point the worker process crashed at the call to cache_free. From the backtrace I could see that the call to cache_free was calling the function with the same name, but in the mod_mem_cache, instead of the local one in the mod_upload_progress. How this was possible I still have no idea (or maybe I didn’t understand the code enough). As soon as I disabled mod_mem_cache, the progress tracking was working like a breeze. It's in the comments of your blog post(http://drogomir.com/blog/2008/6/18/upload-progress-bar-with-mod_passenger-and-apache), and you said you are gonna to fix this issue. As Slobodan advised, I disabled mod_mem_cache, and progress tracking works. I didn't dig into the source code so I have no idea what's wrong.
or cancel

Parsed with GitHub Flavored Markdown

Labels

Apache/Passenger pair returns "Internet Server Error" when uploading large files

Comments

Segmentation fault (11) when mod_upload_progress loaded

Comments

Doens't work with mod_mem_cache

Comments