Author: Padmalatha Somashiandan
Date: March 15, 2013
Bugs you see will typically be one of four types:
Type 1: BPPM panic/crash
Type 2: User/Customer application program
Type 3: Something doesn't work as it should
Type 4: System hangs
Filing a bug (applies to all types of bugs):
============================================
(1) If you are not sure what category/priority to file this bug, talk to some BPPM engineers/bppm95qa@bmc.com - we can help you prioritize the bug and assign the right engineer to look at it.
(ii) Add bppm95qa@bmc.com to the interest list if it is a BPPM bug. If its a bug related to other apps, add NGP_Smoke@bmc.com to the interest list. Similarly for other app specific sub categories, their respective aliases for contact.
(iii) In the description section of the bug report be as precise as possible about the type of problem, what test you were running and what you observed when the bug occurred. Remember to include the following
in the description section:
* Host-name on which the problem was observed.
* Precisely what test you were running:
for e.g. saying "I was accessing CMA while running device related tests" is more useful than saying "Running Device tests with Central Admin".
* What other symptoms you might have seen on the cluster e.g. I/O errors,ProactiveNet.log, ProactiveNetAgent.log, pw process list, agent status, etc.
* If the bug is reproducible, include information on how to reproduce the bug.
* The BPPM version and Solaris/Linux/Windows/x86 version.
Type 1 bugs :
===========================
(a) Check to see if the bug is a duplicate of something that already exists. This may be difficult to do since the same bug can manifest itself in different ways. If you are not sure, file a bug anyway.
(b) A detailed description of the bug; Record whatever information you see on the screen.
A summary of events would be big plus for faster evaluations.
(c) Tell us if the problem is reproducible in clear steps.
(d) If any workarounds. Traces of bppm_server_install_log/proactivenet_server_install/version list/browser type/Java version
Type 2 bugs :
===========
Typically,
(a) Which user/customer application.
(b) What method was used for installation: InstallAnywhere/pkgadd/install.options/AutomationSetup...
(c) Applications integration/configuration details wrt BPPM such
as config.properties, pw p l, pw view history, pw licence list,
state of box before/at the time of hitting the error.
Type 3 bugs (some command doesn't work the way it is supposed to):
====================================================
Be as specific as you can about what the problem
is, the host on which you saw the problem, the time you saw the problem
(so that we can correlate with /var/adm/messages) and what you
were doing on the host before the problem occurred.
Call an specific epic responsible engineer to take a look at the host/box.
Don't reboot/restart any. The engineer looking at the
problem can determine if there is a bug or not.
Type 4 bugs (system hangs):
=====================
The system may hang due to low memory, unusual system activity etc. The
best way to debug these problems is for an engineer to look at the live
host. It is difficult to gather this information from logs. So for
system hangs, call some engineer who can then determine if a bug should
be filed.
Also, while filing bugs, please make sure you collect all the needed
information.
Example : - Console messages
- Log/install list messages
- BPPM version/build, OS/Platform information
Without this information, it is very difficult to debug the problem.
If you need any help with filing bugs, please let us know.
Thanks,
~Latha
Date: March 15, 2013
Bugs you see will typically be one of four types:
Type 1: BPPM panic/crash
Type 2: User/Customer application program
Type 3: Something doesn't work as it should
Type 4: System hangs
Filing a bug (applies to all types of bugs):
============================================
(1) If you are not sure what category/priority to file this bug, talk to some BPPM engineers/bppm95qa@bmc.com - we can help you prioritize the bug and assign the right engineer to look at it.
(ii) Add bppm95qa@bmc.com to the interest list if it is a BPPM bug. If its a bug related to other apps, add NGP_Smoke@bmc.com to the interest list. Similarly for other app specific sub categories, their respective aliases for contact.
(iii) In the description section of the bug report be as precise as possible about the type of problem, what test you were running and what you observed when the bug occurred. Remember to include the following
in the description section:
* Host-name on which the problem was observed.
* Precisely what test you were running:
for e.g. saying "I was accessing CMA while running device related tests" is more useful than saying "Running Device tests with Central Admin".
* What other symptoms you might have seen on the cluster e.g. I/O errors,ProactiveNet.log, ProactiveNetAgent.log, pw process list, agent status, etc.
* If the bug is reproducible, include information on how to reproduce the bug.
* The BPPM version and Solaris/Linux/Windows/x86 version.
Type 1 bugs :
===========================
(a) Check to see if the bug is a duplicate of something that already exists. This may be difficult to do since the same bug can manifest itself in different ways. If you are not sure, file a bug anyway.
(b) A detailed description of the bug; Record whatever information you see on the screen.
A summary of events would be big plus for faster evaluations.
(c) Tell us if the problem is reproducible in clear steps.
(d) If any workarounds. Traces of bppm_server_install_log/proactivenet_server_install/version list/browser type/Java version
Type 2 bugs :
===========
Typically,
(a) Which user/customer application.
(b) What method was used for installation: InstallAnywhere/pkgadd/install.options/AutomationSetup...
(c) Applications integration/configuration details wrt BPPM such
as config.properties, pw p l, pw view history, pw licence list,
state of box before/at the time of hitting the error.
Type 3 bugs (some command doesn't work the way it is supposed to):
====================================================
Be as specific as you can about what the problem
is, the host on which you saw the problem, the time you saw the problem
(so that we can correlate with /var/adm/messages) and what you
were doing on the host before the problem occurred.
Call an specific epic responsible engineer to take a look at the host/box.
Don't reboot/restart any. The engineer looking at the
problem can determine if there is a bug or not.
Type 4 bugs (system hangs):
=====================
The system may hang due to low memory, unusual system activity etc. The
best way to debug these problems is for an engineer to look at the live
host. It is difficult to gather this information from logs. So for
system hangs, call some engineer who can then determine if a bug should
be filed.
Also, while filing bugs, please make sure you collect all the needed
information.
Example : - Console messages
- Log/install list messages
- BPPM version/build, OS/Platform information
Without this information, it is very difficult to debug the problem.
If you need any help with filing bugs, please let us know.
Thanks,
~Latha
No comments:
Post a Comment