RSB Performance Monitoring: Issueshttps://code.cor-lab.de/https://code.cor-lab.de/favicon.ico?14019720732015-04-09T11:28:12ZOpen Source Collaboration Platform
Redmine Bug #2221 (Resolved): NetBandwidthProvider does not notice that a process terminatedhttps://code.cor-lab.de/issues/22212015-04-09T11:28:12ZJ. Wienkejwienke@techfak.uni-bielefeld.de
<p>This would result in a lot of useless processing instead of stopping the thread that monitored the process.</p> Bug #2200 (Resolved): UDP traffic not coveredhttps://code.cor-lab.de/issues/22002015-03-11T16:39:52ZJ. Wienkejwienke@techfak.uni-bielefeld.deBug #2128 (Rejected): memory leak in timeseries adapter with influx backendhttps://code.cor-lab.de/issues/21282014-12-08T10:18:20ZJ. Wienkejwienke@techfak.uni-bielefeld.de
<p>In case influxdb is used as a backend and anything (even completely unrelated things to protocol buffer messages, e.g. manually constructed strings) is written to the data base, then memory grows immediately with protocol buffer internal things accumulating inside the memory:<br /><pre>
Partition of a set of 1624176 objects. Total size = 205469680 bytes.
Index Count % Size % Cumulative % Kind (class / dict of class)
0 145795 9 78640840 38 78640840 38 dict (no owner)
1 142138 9 39798640 19 118439480 58 dict of
google.protobuf.internal.python_message._Listener
2 474935 29 13242264 6 131681744 64 long
3 142138 9 12508144 6 144189888 70 __builtin__.weakproxy
4 195542 12 11513720 6 155703608 76 str
5 94746 6 9853584 5 165557192 81 0x2505270
6 142138 9 9096832 4 174654024 85 google.protobuf.internal.python_message._Listener
7 52629 3 7449256 4 182103280 89 list
8 46004 3 4784416 2 186887696 91 0x2507a70
9 49187 3 3541464 2 190429160 93
google.protobuf.internal.containers.RepeatedCompositeFi
eldContainer
<353 more rows. Type e.g. '_.more' to view.>
Partition of a set of 1624176 objects. Total size = 205470448 bytes.
Index Count % Size % Cumulative % Referrers by Kind (class / dict of class)
0 189492 12 68975088 34 68975088 34 0x2505270
1 142138 9 39798640 19 108773728 53 google.protobuf.internal.python_message._Listener
2 725482 45 26219608 13 134993336 66 dict (no owner)
3 147971 9 15067832 7 150061168 73 list
4 46020 3 12882144 6 162943312 79 0x2507a70
5 142138 9 12508144 6 175451456 85 dict of
google.protobuf.internal.python_message._Listener
6 49187 3 7051736 3 182503192 89
google.protobuf.internal.containers.RepeatedCompositeFi
eldContainer
7 32067 2 2959840 1 185463032 90 types.CodeType
8 45988 3 2943232 1 188406264 92 0x2505270, 0x2507a70,
google.protobuf.internal.containers.RepeatedCompositeFi
eldContainer
9 27054 2 1452928 1 189859192 92 tuple
</pre></p> Bug #2058 (Rejected): Starting the process-monitor with an incorrect PID leads to a strange error...https://code.cor-lab.de/issues/20582014-10-13T18:43:13ZJ. Moringenjmoringe@cor-lab.uni-bielefeld.de
<a name="What-did-I-do"></a>
<h2 >What did I do?<a href="#What-did-I-do" class="wiki-anchor">¶</a></h2>
<pre>
/vol/toolkit/nightly/x86_64/last/bin/rsb-process-monitor0.11 --pid 168 -o /foo
</pre><br />Note that there is no process with id 168.
<a name="What-happened"></a>
<h2 >What happened?<a href="#What-happened" class="wiki-anchor">¶</a></h2>
<pre>
terminate called after throwing an instance of 'std::runtime_error'
what(): Could not determine program name: basic_ios::clear
</pre><br />Note the final word of the error message :)
<a name="What-did-I-expect-to-happen"></a>
<h2 >What did I expect to happen?<a href="#What-did-I-expect-to-happen" class="wiki-anchor">¶</a></h2>
<p>The error should probably contain the process id.</p>
<a name="Notes"></a>
<h2 >Notes<a href="#Notes" class="wiki-anchor">¶</a></h2>
<p>This could be an upstream bug in <a href="https://code.cor-lab.de/projects/rsc" class="project">Robotics Systems Commons</a>.</p> Bug #2057 (Resolved): rsb-process-monitor crashes, probably when target binary has been built for...https://code.cor-lab.de/issues/20572014-10-13T18:39:35ZJ. Moringenjmoringe@cor-lab.uni-bielefeld.de
<a name="What-did-I-do"></a>
<h2 >What did I do?<a href="#What-did-I-do" class="wiki-anchor">¶</a></h2>
<pre>
gdb --args /vol/toolkit/nightly/x86_64/last/bin/rsb-process-monitor0.11 --pid 14854 -o /foo
</pre><br />In this case process 14854 is a Common Lisp logger for i686, i.e. a 32-bit ELF binary.
<a name="What-happened"></a>
<h2 >What happened?<a href="#What-happened" class="wiki-anchor">¶</a></h2>
<pre>
Program received signal SIGSEGV, Segmentation fault.
rsbperfmon::getBuildId (execPath=...) at /home/jenkins/workspace/rsb-performance-monitor-master-toolkit-nightly/label/ubuntu_precise_64bit_techfak/src/common.cpp:174
174 /home/jenkins/workspace/rsb-performance-monitor-master-toolkit-nightly/label/ubuntu_precise_64bit_techfak/src/common.cpp: No such file or directory.
(gdb) bt
#0 rsbperfmon::getBuildId (execPath=...) at /home/jenkins/workspace/rsb-performance-monitor-master-toolkit-nightly/label/ubuntu_precise_64bit_techfak/src/common.cpp:174
#1 0x0000000000453d38 in rsbperfmon::MetaDataProvider::MetaDataProvider (this=0x6a8710, pid=@0x6bafe8: 14854) at /home/jenkins/workspace/rsb-performance-monitor-master-toolkit-nightly/label/ubuntu_precise_64bit_techfak/src/MetaDataProvider.cpp:50
#2 0x00000000004544d1 in rsbperfmon::MetaDataProviderFactory::create (this=<optimized out>, pid=@0x6bafe8: 14854) at /home/jenkins/workspace/rsb-performance-monitor-master-toolkit-nightly/label/ubuntu_precise_64bit_techfak/src/MetaDataProvider.cpp:70
#3 0x0000000000455b4a in rsbperfmon::MonitoringThread::addCueProvider (this=0x6baec0, factory=...) at /home/jenkins/workspace/rsb-performance-monitor-master-toolkit-nightly/label/ubuntu_precise_64bit_techfak/src/MonitoringThread.cpp:70
#4 0x000000000042e546 in main (argc=<optimized out>, argv=<optimized out>) at /home/jenkins/workspace/rsb-performance-monitor-master-toolkit-nightly/label/ubuntu_precise_64bit_techfak/src/process-monitor.cpp:244
</pre>
<a name="What-did-I-expect"></a>
<h2 >What did I expect?<a href="#What-did-I-expect" class="wiki-anchor">¶</a></h2>
<p>The monitor should either display a proper error message, successfully monitor the binary or monitor the binary with a reduced set of cues.</p> Feature #1889 (Resolved): Allow thread monitoring in combination with subprocess monitoringhttps://code.cor-lab.de/issues/18892014-05-26T15:51:20ZJ. Wienkejwienke@techfak.uni-bielefeld.de
<p>For the robocup system, a lot of programs are based on java and hence have a bash-wrapper around them. It would be nice to monitor their threads, which is currently not possible because subprocess monitoring is required to traverse to the java subprocess and subprocess monitoring prevents thread monitoring.</p>