Client error
31 Mar 2008 18:06:25 UTC
Topic 193597
(moderation:
Thats the message i am getting in 6 different workunits since some 4 weeks, looking up the co cruncher i see they have the same problems, so deducting thats a problem with the workunit and not with the computer. as I sum some 50 hours of computing could i kindly ask for a fix of this or should i quit computing Einstein until you discover the reason of the client errors?
best regards Nikolaus
zip error: Could not create output file (../../projects/einstein.phys.uwm.edu/h1_0859.65_S5R3__241_S5R3b_3_0)
2008-03-31 13:48:06.3917 [normal]: WARNING: Can't zip output file 'h1_0859.65_S5R3__241_S5R3b_3_0'
2008-03-31 13:48:06.3923 [normal]: done. calling boinc_finish(0).
called boinc_finish
h1_0859.65_S5R3__241_S5R3b_3_0
-161
]]>
Validate state Inva
Client error
)
I've looked at your results list and indeed there are several (actually 7) client errors currently showing. It's not a problem with the task because most of the other computers you are paired up with aren't having a problem and the few that are, are having quite a different problem.
The unfortunate thing is that all tasks are being crunched to completion and quite successfully by the look of things. Here are the actual messages from the stderr.txt output for the latest client error:-
.... 1194, 1195, 1196, c 1197, 1198, 1199, c 1200, done. FPU status flags: COND_3 COND_0 PRECISION zip I/O error: Interrupted system callzip error: Could not create output file (../../projects/einstein.phys.uwm.edu/h1_0859.65_S5R3__241_S5R3b_3_0)
2008-03-31 13:48:06.3917 [normal]: WARNING: Can't zip output file 'h1_0859.65_S5R3__241_S5R3b_3_0'
2008-03-31 13:48:06.3923 [normal]: done. calling boinc_finish(0).
called boinc_finish
h1_0859.65_S5R3__241_S5R3b_3_0
-161
In the above, I've highlighted in blue, the successful completion of crunching. What happens next is that the output of crunching the task is supposed to be zipped up and uploaded back to the project. It would appear that the implementation of the zip function in BOINC is having intermittent problems making a system call on your machine. Most of the time things work but about 10% of the time you get a zip I/O error. Seems rather weird. Maybe someone who knows something about Darwin might be able to give us a better appraisal. It sort of looks a bit like a problem with a system library perhaps.
Cheers,
Gary.
RE: RE: Thats the message
)
Thanks a lot for this insight, now who is gone a save my poor crunched units from being thrown away or re phrased what can i do to avoid this unnecessary losses of crunched units for the einstein, is there any einstein out there who can help'
RE: RE: RE: Thats the
)
Errors keep popping up
well will leave einstein until you guys get these issues solved
RE: .... 1194, 1195, 1196,
)
...is it possible that you are running out of disk space?
some WUs get zipped and reported correctly, but sometimes ZIP get interrupted.
..or do you have an virus scanner which might interrupt ZIP?
Udo
Udo
Hello to all, I got this
)
Hello to all,
I got this -161 error one one of my linux hosts (openSUSE 10.2) constantly durign the last days with every workunit.
snippet from one example workunit:
------------
1190, 1191, 1192, c
1193, 1194, 1195, c
1196, 1197, 1198, c
1199, 1200, 1201, c
done.
FPU status flags: PRECISION
2008-04-05 12:04:56.5961 [normal]: done. calling boinc_finish(0).
called boinc_finish
h1_0998.30_S5R3__589_S5R3b_1_0
-161
]]>
-----------------
Seems the calculation was finished sucessfully.
This particular host runs BOINC 6.14 (built from svn), but that could't be the problem, because I have 2 other hosts running this version without any problem.
I will try to go back to the official release version. Perhaps it helps.
RE: Hello to all, I got
)
Hi Mahdia, if I am reading this correctly then no output file was generated on completion of the WU and there was no result for your BOINC client to upload to the project. A check of the messages tab will confirm if I am correct. I am afraid that I cannot help you as to why this should fail on only one of your hosts but 6.1.14 is the current developmental version and causes this exact error (at least in the Mac flavour) with Einstein@home. A fix is currently being explored. Dave.
RE: I cannot help you as
)
After posting my message I read in another thread about this problem.
I changed the client version back to 5.10.45. Perhaps this will help...
I got error -161 running
)
I got error -161 running boinc_ubuntu_6.1.14_i686-pc-linux
and einstein_S5R3_4.38_i686-pc-linux.
After that error I changed the client version back to 5.10.45!
RE: I got error -161
)
Error -161 is what you get when BOINC cannot find the task, either because it doesn't exist in the correct place, or because BOINC can't find client_state.xml
Now the thing is, 6.1.14 uses sandboxing and needs a separate BOINC and BOINC Data directory. If all your data and appropriate files aren't migrating to the BOINC Data directory, you get errors like that. 6.1 is still in testing though, so if you do not want to test out the migration process, stick with the stock and stable 5.10.45 until BOINC 6.2 stable comes out.
No more error -161 when using
)
No more error -161 when using client boinc_ubuntu_6.1.15_i686-pc-linux and einstein_S5R3_4.38_i686-pc-linux!