obmcutil is slow, nonsensical output
root@palmetto:~# time obmcutil 2>&1 > /dev/null
real 0m3.301s user 0m3.030s sys 0m0.230s
root@palmetto:~# obmcutil poweron = None
The above is really confusing as to what is actually going on... does that mean success? Does that mean there's no power? Does that mean no action was taken?
Signature · 2 lines
Stewart Smith OPAL Architect, IBM.
We see the same behavior on the Barreleye servers, the obmcutil is quite slow but we've gotten used to it. The poweron and poweroff commands usually return either this message or a python backtrace if the backend dbus services are still warming up.
________________________________________
From: openbmc <[email protected]> on behalf of Stewart Smith <[email protected]>
Sent: Thursday, March 31, 2016 1:02 AM
To: openbmc
Subject: obmcutil is slow, nonsensical output
root@palmetto:~# time obmcutil 2>&1 > /dev/null
real 0m3.301s user 0m3.030s sys 0m0.230s
root@palmetto:~# obmcutil poweron = None
The above is really confusing as to what is actually going on... does that mean success? Does that mean there's no power? Does that mean no action was taken?
--
Stewart Smith
OPAL Architect, IBM.
_______________________________________________
openbmc mailing list
[email protected]
https://lists.ozlabs.org/listinfo/openbmc
obmcutil was intended as a lab debug tool only. It is simply a lookup of shortcuts to the dbus calls. It prints out whatever the dbus call returns. All of the slowness is due to python importing several libraries which is
slow because the CPU is already highly utilized. Brad looked into this at
one point and collected some data.
Regards,
Norman James
IBM - POWER Systems Architect
Phone: 1-512-286-6807 (T/L: 363-6807)
Internet: [email protected]
From: Kenneth Wilke <[email protected]>
To: Stewart Smith <[email protected]>, openbmc
<[email protected]>
Date: 03/31/2016 10:35 PM
Subject: Re: obmcutil is slow, nonsensical output
Sent by: "openbmc" <[email protected]>
We see the same behavior on the Barreleye servers, the obmcutil is quite slow but we've gotten used to it. The poweron and poweroff commands usually return either this message or a python backtrace if the backend dbus services are still warming up.
________________________________________
From: openbmc <openbmc-bounces [email protected]> on behalf of Stewart Smith <[email protected]>
Sent: Thursday, March 31, 2016 1:02 AM
To: openbmc
Subject: obmcutil is slow, nonsensical output
root@palmetto:~# time obmcutil 2>&1 > /dev/null
real 0m3.301s user 0m3.030s sys 0m0.230s
root@palmetto:~# obmcutil poweron = None
The above is really confusing as to what is actually going on... does that mean success? Does that mean there's no power? Does that mean no action was taken?
--
Stewart Smith
OPAL Architect, IBM.
_______________________________________________
openbmc mailing list
[email protected]
https://lists.ozlabs.org/listinfo/openbmc
_______________________________________________
openbmc mailing list
[email protected]
https://lists.ozlabs.org/listinfo/openbmc
Quoted 4 lines · Kenneth Wilke
We see the same behavior on the Barreleye servers, the obmcutil is quite slow but we've gotten used to it. The poweron and poweroff commands usually return either this message or a python backtrace if the backend dbus services are still warming up.
Python backtrace isn't very elegant :(
Signature · 2 lines
Stewart Smith OPAL Architect, IBM.