GDP

09/28/2016

10:25 PM Bug #54: Multiple multi-reads (of size 1, typically) do not return the correct number of events.

commit:4d6ff457 seems to have it fixed. I can no longer reproduce the bug.
Leaving this open though, just in case... Nitesh Mor

10:22 PM Bug #59 (Closed): Segmentation fault in gdp-reader on subscriptions

On a log with no new appends, leaving a subscription running causes segmentation fault after a while. See a detailed ... Nitesh Mor

09:50 PM Bug #55: GDP log subscribe hangs

`__func1` is just a name for `gdp_event_next`. As per the trace in the JPEG file (really?), it is called with a NULL ... Nitesh Mor

05:16 PM Bug #58: 403 forbidden errors for many consecutive asynchronous writes (15 bytes @ 100 Hz)

At the moment, this is happening reasonably often for data at 10 Hz. Weirdly, ping times seem to be better:
~~~
P... David Mellis

09/27/2016

04:52 PM Bug #57: GDP Make Error

I just tried gdp-setup.sh.
On Raspberry Pi it doesn't work because Raspbian is not supported by the script. Does thi... Nima Mousavi

12:54 PM Bug #57 (In Progress): GDP Make Error

Have you tried running adm/gdp-setup.sh? The recent query-by-timestamp function has added a dependency on libdb. Eric Allman

09:59 AM Bug #57 (Closed): GDP Make Error

I have pulled the most recent versions of GDP, but when I run sudo make, I receive a fatal error. I've attached the e... Nima Mousavi

04:07 PM Bug #58 (Closed): 403 forbidden errors for many consecutive asynchronous writes (15 bytes @ 100 Hz)

I'm piping data of around 15 bytes per reading (line) at 100 Hz to `./gdp-writer -a edu.berkeley.eecs.bid.mellis.ardu... David Mellis

11:25 AM Bug #56: Crash if isatty() returns true but TERM isn't set.

Thanks! David Mellis

07:07 AM Bug #56 (Resolved): Crash if isatty() returns true but TERM isn't set.

Sorry about that. Should be fixed in commit:0decb227. Eric Allman

07:22 AM Bug #55: GDP log subscribe hangs

Nitesh, I'm passing this one to you because Nima is seeing the problem in __func1, which is part of the Python API. ... Eric Allman

07:21 AM Bug #49 (Closed): Darwin: shared libraries have the wrong internal name

Thanks, this works for me now so I'm closing this. Anonymous

07:14 AM Bug #50: router will sometimes connect to itself

Dropping this to low priority because at the moment it doesn't seem to be causing any operational problems.
Might ... Eric Allman

09/26/2016

02:04 PM Bug #56: Crash if isatty() returns true but TERM isn't set.

BTW, this was for an old commit (202d751a0d47c3ecd6d025e68b4ddb4f7a665b45) so it may not apply any more. David Mellis

01:33 PM Bug #56 (Closed): Crash if isatty() returns true but TERM isn't set.

I'm getting a crash on line 289 of ep_string.c when running an application linked against the GDP from within Xcode. ... David Mellis

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

I've been using the function calls that have been used in gdp/lang/python/reader_multi_subscribe.py, and I'm facing t... Nima Mousavi

09/24/2016

07:30 PM Bug #54 (Closed): Multiple multi-reads (of size 1, typically) do not return the correct number of...

For a multi-read of size 1, a client should receive 2 events (GDP_EVENT_DATA followed by GDP_EVENT_EOS). In case of N... Nitesh Mor

09/23/2016

10:06 AM Bug #53 (Closed): Segfault in gdp-reader.c; cleanup of signature buffer

Seems to work. Closing. Nitesh Mor

02:22 AM Bug #53 (Feedback): Segfault in gdp-reader.c; cleanup of signature buffer

The analysis is correct; the signature buffer reuse was broken. Fixed in commit:354ee62a.
We should really elimin... Eric Allman

01:10 AM Bug #53 (Closed): Segfault in gdp-reader.c; cleanup of signature buffer

In the case of a log with signatures, an EOF causes a segmentation fault. Here's an example of how to reproduce it:
... Nitesh Mor

12:45 AM Bug #52 (Closed): Repeated gdp_gcl_open calls on non-existing GCLs do not give proper error code ...

Seems to work. Closing the issue. Nitesh Mor

12:14 AM Bug #52 (Feedback): Repeated gdp_gcl_open calls on non-existing GCLs do not give proper error cod...

Problem was that the unopened GCL was being kept in the cache; the subsequent calls to gdp_gcl_open found the cached ... Eric Allman

09/19/2016

09:18 PM Bug #52: Repeated gdp_gcl_open calls on non-existing GCLs do not give proper error code (was: GDP...

Good point. There was a minor change in `GDP_GCL(...)` to better match it with the underlying C-API behavior. More sp... Nitesh Mor

09/18/2016

05:42 AM Bug #52 (Closed): Repeated gdp_gcl_open calls on non-existing GCLs do not give proper error code ...

When opening a non-existing GCL, the gdp.GDP_GCL function in the Python API only raises an error during the first att... Rico Maestro

09/09/2016

02:43 PM Feature #46 (Feedback): Need way of rebuilding log indices ("fsck for logs")

There's a version of gdp-log-check in the apps directory that at least does the basics. I'm putting this into Feedba... Eric Allman

11:44 AM Bug #49: Darwin: shared libraries have the wrong internal name

These issues should all be fixed now. As it turns out libs was never being cleaned at all. Eric Allman

09/07/2016

02:18 PM Bug #35 (Closed): EP Crypto Error

Nitesh Mor

11:11 AM Bug #35 (Resolved): EP Crypto Error

Nitesh Mor

02:17 PM Bug #36 (Closed): Error 409

Nitesh Mor

02:17 PM Bug #29 (Closed): Internal Server Error

Nitesh Mor

11:06 AM Bug #21 (Closed): "500: internal server error" in the visualization tool.

Nitesh Mor

11:05 AM Bug #28 (Closed): gdplogd losing connection to the router. signature write failure: Success

Nitesh Mor

11:05 AM Bug #38 (Closed): Resource exhaustion after repeated router restarts

Nitesh Mor

11:04 AM Bug #41 (Closed): gdp-writer segfaults under Mac OS X

Nitesh Mor

11:03 AM Bug #47 (Closed): No sync to disk after timestamp index update

Doing a sync at the time of closing and a potential recovery operation before/during gdplogd startup seems to be a go... Nitesh Mor

09/06/2016

04:52 PM Bug #49: Darwin: shared libraries have the wrong internal name

Thanks for making that fix, the path name is in the library is correct under Mac OS X and the binary that is created ... Anonymous

04:31 PM Bug #49 (Feedback): Darwin: shared libraries have the wrong internal name

I have changed libs/makelinks.sh to do the
install_name_tool -id libgdp.0.7.dylib ../libs/libgdp.0.7.dylib
... Eric Allman

02:57 PM Bug #49: Darwin: shared libraries have the wrong internal name

This bug has not been fixed, so it should not be closed. Creating a .so file instead of a .dylib file causes problem... Anonymous

02:43 PM Bug #49 (Closed): Darwin: shared libraries have the wrong internal name

Nitesh Mor

09/05/2016

04:23 PM Bug #16 (Closed): Threaded GDP Clients Fail

Verified that it works from Python API as well. Marking as closed. Nitesh Mor

09/04/2016

04:10 PM Bug #16: Threaded GDP Clients Fail

I've added an updated version of stuck-sub.c that gives more information, such as the return status of various GDP AP... Eric Allman

03:23 PM Bug #16 (Resolved): Threaded GDP Clients Fail

I believe this is fixed by commit:1aa4fc2c. The problem was that the lack of a lock around the open calls permitted ... Eric Allman

09/02/2016

04:44 PM Bug #16 (In Progress): Threaded GDP Clients Fail

I've found some problems, and it behaves differently now, but it still isn't correct. Some of the problems may be in... Eric Allman

04:39 PM Feature #46 (In Progress): Need way of rebuilding log indices ("fsck for logs")

A version of `gdp-log-check` has been checked in, but it isn't well tested yet. I recommend against trying it in pro... Eric Allman

08/31/2016

01:04 PM Bug #16: Threaded GDP Clients Fail

Reproduced with C library. See attached sample program, here's the output (ignore the timestamp precision, for the mo... Nitesh Mor

08/30/2016

10:19 AM Bug #16: Threaded GDP Clients Fail

I think I can reproduce this bug now. At a first look, this does not appear to be specific to the Python API, but nee... Nitesh Mor