Dear Antun,
Sorry for answering only now. Piotr is away this and
the last weeks. I'm taking care for SFT for that time.
The 'spelling error' corrected in CVS, was another problem,
not this one. This appears in case the command you mentioned
count=`rgma -c "select count(*) from LcgRecords where
ExecutingCE='$SFT_NODE_NAME' and MeasurementDate>='$date'"
doesn't return a result (in the expected format)
+----------+
| count(*) |
+----------+
| 28 |
+----------+
1 Rows in set
I'm correcting this bug, so just the relevant error message
would appear.
Judit
On h, aug 29, Antun Balaz wrote:
> Hi,
> nobody seems to mind (at least nobody bothered to answer my previous
> question along the same lines), but I think that accounting is very
> important, so I will rise this question again...
>
> I have managed to fix APEL at my site, i.e. when I issue command like
>
> rgma -c 'select count(*) from LcgRecords where
> ExecutingCE='\''ce.phy.bg.ac.yu'\'' and MeasurementDate>='\''2005-08-28'\'''
>
> it is always successful and returns nonzero count. The same applies to SFT2
> tests submitted by me.
>
> However, APEL tests from officially submitted SFT2 tests can still fail from
> time to time, with the cryptic error
>
> Count:
> tests/sft-apel: line 28: [: -gt: unary operator expected Empty Apel result
> set - FAILED!
>
> Details at
>
> https://lcg-sft.cern.ch:9443/sft/sitehistory.cgi?site=ce.phy.bg.ac.yu
>
> clicking on different APEL test results.
>
> I suspect that the problem lies with the use of earlier versions of SFT2
> software from time to time, since history log of SFT2 mentions "fixed
> spelling error" for sft-apel test.
>
> Can somebody take a look at this, please?
>
> Thanks, Antun
>
> -----
> E-mail: [log in to unmask]
> Web: http://scl.phy.bg.ac.yu/
>
> Phone: +381 11 3160260, Ext. 152
> Fax: +381 11 3162190
>
> Scientific Computing Laboratory
> Institute of Physics, Belgrade
> Serbia and Montenegro
> -----
|