GDP

11/19/2016

06:45 AM Bug #75 (Closed): Subscription times out

Rico, I am relieved that the problem's been tracked down. Good sleuthing job, and thanks for letting us know. Eric Allman

09:19 PM Bug #75: Subscription times out

Hi Eric,
After rereading the documentations for installing and running the log server, and checking if I've follow... Rico Maestro

11/14/2016

11:47 AM Bug #82: Memory leaks in subscriptions (and elsewhere)

Thanks for taking a look, I agree that these leaks are of low priority.
In an embedded system a process could run ... Anonymous

11:12 AM Bug #82: Memory leaks in subscriptions (and elsewhere)

There are many cases where memory is allocated during startup and never released; for example, when reading the admin... Eric Allman

07:44 AM Bug #82: Memory leaks in subscriptions (and elsewhere)

I can still reproduce this under RHEL with today's version of the GDP.
~~~
[cxh@terra apps]$ gcl-create -k none p... Anonymous

11:21 AM Bug #88 (New): memory leak in GCL cache hash table

The open GCL cache (as implemented in `gdp/gdp_gcl_cache.c`) uses `libep` hash tables (`ep/ep_hash.c`). They in turn... Eric Allman

11/13/2016

01:09 PM Berkeley servers (gdp-01, -02, -03, and -04) updated

The four GDP servers at Berkeley have been updated to the latest development version of the system. This is _not_ an... Eric Allman

11/12/2016

03:31 PM Bug #86 (Closed): Multiple subscriptions and/or multireads to the same log don't work

Seems to work. I can no longer reproduce this bug. Closing. Nitesh Mor

02:14 PM Bug #69 (Feedback): Read, Write Errors

Syed, could you try this again, preferably using the latest `gdplogd` from the repo? I think it may be fixed, and I ... Eric Allman

02:10 PM Bug #77 (Resolved): GDP Log Server Crashes when we end subscription

It looks like this is the same problem as #78, which is closed. Syed, could you see if this is fixed? You may need ... Eric Allman

02:05 PM Bug #84 (Resolved): Assertion in gdp_gcl_create() crashes the calling process

Eric Allman

02:02 PM Bug #75: Subscription times out

I'm completely stumped — I can't reproduce this at all, at least using the current code (I tried both the latest `gdp... Eric Allman

10:26 AM Bug #75: Subscription times out

Hi Eric. Here is the output of the command in our server.
`@(#)gdplogd 0.7.2 (2016-11-12 11:53) 4f976ad9f75dad291b... Rico Maestro

09:55 AM Bug #75: Subscription times out

Rico, I'm not able to reproduce this on my latest code version (and yes, there have been some changes to the way subs... Eric Allman

02:18 AM Bug #75: Subscription times out

Hi Nitesh,
I have compiled the latest version of the gdplogd codes from the repo and tried if the bug is already f... Rico Maestro

10:18 AM Bug #40 (Resolved): gcl-create calls abort: there should be a version of _gdp_gcl_decref that doe...

I believe this is related to (or the same as) issue #83, which has been fixed. Eric Allman

04:25 PM Feature #46 (Closed): Need way of rebuilding log indices ("fsck for logs")

I'm closing this issue since I think a preliminary program is available that works acceptably. Enhancements should b... Eric Allman

11/11/2016

03:07 PM Bug #83 (Closed): Assertions should not crash the calling process: assertion in gdp_gcl_close() c...

Cool, thanks, I think we can close this one, so I'm closing it. Anonymous

11:19 AM Bug #83: Assertions should not crash the calling process: assertion in gdp_gcl_close() causes the...

As of commit:8701c522 `gdp_gcl_close` should return `GDP_STAT_GCL_NOT_OPEN` when passed a bad GCL. Eric Allman

02:51 PM Bug #80 (Closed): GDP setup

Closed due to lack of feedback. Eric Allman

02:49 PM Bug #82 (Feedback): Memory leaks in subscriptions (and elsewhere)

I can't seem to reproduce this on the latest version (although admittedly I'm testing on MacOS; Linux may be differen... Eric Allman

02:35 PM Bug #84: Assertion in gdp_gcl_create() crashes the calling process

I believe this assertion has been fixed. Since it's a "can't happen" assertion it's hard to reproduce. None the les... Eric Allman

10:12 AM Bug #86 (Resolved): Multiple subscriptions and/or multireads to the same log don't work

I believe this should be fixed in commit:029cf007. It turns out it was in gdplogd, not libgdp. Eric Allman

11/08/2016

02:54 PM Bug #70: Makefiles do not respect environmental CFLAGS

These other problems should have been resolved in commit:6cc67fbe. Eric Allman

02:47 PM Bug #87 (Resolved): Crash when trying to use RSA key to sign a log

Should be fixed. There was an undersized buffer. Note that this was not a security problem: no buffer overflows occ... Eric Allman

02:04 PM Bug #87 (Closed): Crash when trying to use RSA key to sign a log

~~~
vagrant@vagrant-ubuntu-trusty-64:~$ gcl-create -k rsa -K test_keys/ -s com.pbramsen.gdplogd my_rsa_log
Enter PE... Paul Bramsen

11/03/2016

09:21 AM Bug #83: Assertions should not crash the calling process: assertion in gdp_gcl_close() causes the...

The problem seems to be that if I run the model twice, then it sometimes crashes because close() is called twice.
... Anonymous

11/01/2016

01:58 PM Bug #71: Python shared library version mismatch

It is possible to build without avahi by using "make all_noavahi". Mosquitto is not typically required at runtime, o... Anonymous

01:37 PM Bug #71: Python shared library version mismatch

As near as I can tell, homebrew versus macports is kind of a toss-up, since neither supersets the other. In particul... Eric Allman

10:25 AM Bug #71: Python shared library version mismatch

Nitesh Mor wrote:
> Yepp, seems to have been fixed in commit:152bd9. Thanks Christopher.
Awesome, thanks.
I no... Pat Pannuto

10:20 AM Bug #71 (Closed): Python shared library version mismatch

Yepp, seems to have been fixed in commit:152bd9. Thanks Christopher. Nitesh Mor

09:09 AM Bug #71: Python shared library version mismatch

I may have fixed this, but Nitesh should confirm. Anonymous

01:42 PM Bug #78 (Closed): A routing failure causes gdplogd to become non-responsive.

Nitesh Mor

10:41 AM Bug #86 (Closed): Multiple subscriptions and/or multireads to the same log don't work

*This is a long known bug, but hadn't been reported earlier. Reporting it to make sure everyone is aware of it and du... Nitesh Mor

10:36 AM Bug #70: Makefiles do not respect environmental CFLAGS

Eric Allman wrote:
> Fixed in commit:f5c1552e. Sorry something this simple took so long. Please verify that it's f... Pat Pannuto

06:49 PM Bug #70 (Resolved): Makefiles do not respect environmental CFLAGS

Fixed in commit:f5c1552e. Sorry something this simple took so long. Please verify that it's fixed for you. Eric Allman

10:23 AM Bug #55 (Closed): GDP log subscribe hangs

Closing this, since it's a different bug (multiple subscriptions and/or multi-reads to the same log in a single clien... Nitesh Mor

10/31/2016

01:35 PM Bug #84: Assertion in gdp_gcl_create() crashes the calling process

Your version that deals more forgivingly with assertions sounds like a good thing.
Having preconditions is a good ... Anonymous

01:00 PM Bug #84: Assertion in gdp_gcl_create() crashes the calling process

I don't think you appreciate that assertions are not bugs themselves, they are symptoms of bugs. Feel free to create... Eric Allman

08:58 AM Bug #84 (In Progress): Assertion in gdp_gcl_create() crashes the calling process

If people are to use the the GDP library, then it must be robust. Part of that robustness is to properly handle any ... Anonymous

09:06 AM Support #85 (Closed): Create a GDP public release

The GDP seems fairly close to being releasable to the public.
Perhaps the SwarmLab meeting on November 15 would be... Anonymous

10/28/2016

01:03 PM Support #79 (Feedback): Async write/read

More detail, please. Are the async appends sent from a single thread or multiple threads? Are all records corrupted... Eric Allman

12:11 PM Bug #83 (Feedback): Assertions should not crash the calling process: assertion in gdp_gcl_close()...

I'm not 100% convinced that this is actually a duplicate of Issue #84, since it happened on a call to `gdp_gcl_close`... Eric Allman

12:01 PM Bug #84 (Feedback): Assertion in gdp_gcl_create() crashes the calling process

This is not a bug. If you look carefully, `gdp_init` failed:
gdp_init: ERROR: Network is unreachable [EPLIB:er... Eric Allman

10/25/2016

06:43 PM Bug #55: GDP log subscribe hangs

Well that explains it then.
I'm now just using subscribe and it seems to be working.
You may want to close this bug... Nima Mousavi

10/24/2016

12:49 AM Bug #55: GDP log subscribe hangs

Just a quick update, gdp-01 has been upgraded to the same state as the other three servers. So, you should not have t... Nitesh Mor