If you have Blahd records it use them instead of the GkRecords and
MsgRecords tables. This is used in a LCG CE with patch 898.
This site has published new accounting data:
rgma> select ExecutingSite, MeasurementDate, count(*), min(EventDate),
max(EventDate) from LcgRecords where ExecutingSite LIKE 'CY-01-KIMON'
group by 2,1
+---------------+-----------------+----------+----------------+----------------+
| ExecutingSite | MeasurementDate | count(*) | min(EventDate) |
max(EventDate) |
+---------------+-----------------+----------+----------------+----------------+
| CY-01-KIMON | 2007-12-20 | 7367 | 2007-11-18 | 2007-12-20 |
+---------------+-----------------+----------+----------------+----------------+
Regards,
Pablo
On 20/12/2007 10:30, Asterios Katsifodimos wrote:
> Hello again,
>
> Now, the data go to the EventsRecords database into the MON.
> The problem now is that the records from 2007-11-15 until now
> are in the EventsRecords but the GkRecords table does not have
> any entries from 2007-11-15 until today.
>
> This is due to the reason that GkProccessor was disabled from the APEL
> configuration file on our CE.
> I enabled it and it produced many entries that were not created from
> that date.
>
> Is this supposed to be disabled?
>
> I saw in APEL's code that the APEL publisher makes a JOIN
> of the GkRecords, EventsRecords,SpecRecords and MsgRecords tables.
> So, if any of them does not have tuples for some given dates(e.g.
> GkRecords), then
> the JOIN produces emplty recordsets(this is the case with us).
>
>
> Can someone check if data are moved to the
> http://www3.egee.cesga.es/gridsite/accounting/CESGA
> <http://www3.egee.cesga.es/gridsite/accounting/CESGA>
> accounting database?
>
> thanks,
> Asterios
>
> On Dec 19, 2007 11:28 AM, Asterios Katsifodimos
> <[log in to unmask] <mailto:[log in to unmask]>> wrote:
>
> Hello,
>
> I searched all the log files with no luck.
> The inspectTable*s* variable is ok in the config file.
>
> The rgma-server-check can communicate with the registry.
>
> What I saw (sniffing the packets that are sent from the CE to the
> MON when
> the/opt/glite/bin/apel-pbs-log-parser script runs )is this:
>
> 2955.039959 194.42.27.253(*CE*) -> 194.42.27.251(*MON* box) MySQL
> Request Command: Unknown (23) :
> \001\000\000\000\000\001\000\000\000\000\000\000\001\375\000\375\000\375\000\375\000\375\000\375\000\375\000\375\000\375\000\375\000\375\000\375\000\375\000\375\000\375\000\375\000\375\000\375\000\375\000\375\000\375\000P2007-12-19
> 10:19:43 241858.ce101.grid.ucy.ac.cy
> <http://241858.ce101.grid.ucy.ac.cy> ce101.grid.ucy.ac.cy
> <http://ce101.grid.ucy.ac.cy>
> CY-01-KIMON\vCY-01-KIMON\033241858.ce101.grid.ucy.ac.cy\tbiomed014\006biomed\006P1M34S\003P8S\00294\0018\0242007-12-19T08:18:09Z\0242007-12-19T08:19:43Z\0242007-12-19T08:18:09Z\0242007-12-19T08:19:43Z\n1198052289\n1198052383\024ce101.grid.ucy.ac.cy\00516300\00578204\0010\n2007-12-19\b10:19:43
>
> 2955.040662 194.42.27.251 <http://194.42.27.251> -> 194.42.27.253
> <http://194.42.27.253> MySQL Response Error Code: 426
>
>
> I googled for error code 426 and 1426 and I found this:
> *ERROR 1426:* Message: Too big precision %d specified for column
> '%s'. Maximum is %d.
>
> So, the CE cannot send the data to the MON box. Could you suspect
> anything in this?
>
>
> thanks again,
>
>
> On Dec 19, 2007 12:17 AM, Kyriakos Ginis <
> [log in to unmask]
> <mailto:[log in to unmask]>> wrote:
>
> On Tue, Dec 18, 2007 at 05:51:55PM +0200, Asterios
> Katsifodimos wrote:
> > Hello,
> >
> > On the mon box I can see this for today:
> > Tue Dec 18 04:38:07 UTC 2007: apel-publisher -
> > ====================================
> > Tue Dec 18 04:38:07 UTC 2007: apel-publisher -
> Synchronisation data
> > check
> > Tue Dec 18 04:38:07 UTC 2007: apel-publisher -
> > ====================================
> > Tue Dec 18 04:38:07 UTC 2007: apel-publisher - Finding all
> records in local
> > database since the last successful publish timestamp :
> 2007-11-24 05:22:48
> > Tue Dec 18 04:38:08 UTC 2007: apel-publisher - No records found
> > Tue Dec 18 04:38:08 UTC 2007: apel-publisher -
> > ====================================
> > Tue Dec 18 04:38:08 UTC 2007: apel-publisher - Completed
> Synchronisation
> > data check
> > Tue Dec 18 04:38:08 UTC 2007: apel-publisher -
> > ====================================
> > Tue Dec 18 04:38:08 UTC 2007: apel-publisher - Publisher
> Mode = Apel
> > Publisher (Default)
> > Tue Dec 18 04:38:08 UTC 2007: apel-publisher - Building
> account records via
> > the new Accounting Log File
> > Tue Dec 18 04:38:08 UTC 2007: apel-publisher - NB: Record
> Counts may be zero
> > if Patch #898 is not active on this CE
> > Tue Dec 18 04:38:08 UTC 2007: apel-publisher - Stitching
> together all
> > accounting records
> > Tue Dec 18 04:38:08 UTC 2007: apel-publisher - Stitching
> completed
> > Tue Dec 18 04:38:08 UTC 2007: apel-publisher - No accounting
> data to store
> > Tue Dec 18 04:38:08 UTC 2007: apel-publisher - Number of
> Joined accounting
> > records: 0
> > Tue Dec 18 04:38:08 UTC 2007: apel-publisher - Build complete
> > Tue Dec 18 04:38:08 UTC 2007: apel-publisher - Building
> account records via
> > GK Logs
> > Tue Dec 18 04:38:08 UTC 2007: apel-publisher - NB: Record
> Counts may be zero
> > if Patch #898 is active on this CE
> > Tue Dec 18 04:38:08 UTC 2007: apel-publisher - Stitching
> together all
> > accounting records
> > Tue Dec 18 04:38:10 UTC 2007: apel-publisher - Stitching
> completed
> > Tue Dec 18 04:38:10 UTC 2007: apel-publisher - No accounting
> data to store
> > Tue Dec 18 04:38:10 UTC 2007: apel-publisher - Number of
> Joined accounting
> > records: 0
> > Tue Dec 18 04:38:10 UTC 2007: apel-publisher - Build complete
> > Tue Dec 18 04:38:10 UTC 2007: apel-publisher - Publishing
> data into rgma
> > Tue Dec 18 04:38:10 UTC 2007: apel-publisher - Publishing
> Records to GOC
> > (via Accounting Log): 0
> > Tue Dec 18 04:38:10 UTC 2007: apel-publisher - Publishing
> data into rgma
> > Tue Dec 18 04:38:10 UTC 2007: apel-publisher - Publishing
> Records to GOC
> > (via GK Log): 0
> > Tue Dec 18 04:38:10 UTC 2007: apel-publisher - **** Join
> processing complete
> > ****
> > Tue Dec 18 04:38:10 UTC 2007: apel-publisher -
> > ====================================
> > Tue Dec 18 04:38:10 UTC 2007: apel-publisher -
> Publishing Summary
> > Data
> > Tue Dec 18 04:38:10 UTC 2007: apel-publisher -
> > ====================================
> > Tue Dec 18 04:38:10 UTC 2007: apel-publisher - Data will be
> written to RGMA
> > Table : LcgRecordsSync_v2
> > Tue Dec 18 04:38:10 UTC 2007: apel-publisher - Creating a
> new Primary
> > Producer
> > Tue Dec 18 04:38:14 UTC 2007: apel-publisher - Publishing
> summary data into
> > rgma
> > Tue Dec 18 04:38:19 UTC 2007: apel-publisher - ------
> Processing finished
> > ------
> >
> > This means that the mon box had nothing to publish.
> > However, the CE says:
> > Tue Dec 18 13:44:34 UTC 2007: apel-pbs-log-parser - Event
> records inserted:
> > 364
> > Tue Dec 18 13:44:34 UTC 2007: apel-pbs-log-parser - Checking
> the
> > BlahdRecords table
> > Tue Dec 18 13:44:34 UTC 2007: apel-pbs-log-parser - The
> BlahdRecords schema
> > is up-to-date
> > Tue Dec 18 13:44:34 UTC 2007: apel-pbs-log-parser -
> Reprocess disabled,
> > checking new event logs only
> > Tue Dec 18 13:44:35 UTC 2007: apel-pbs-log-parser - Blahd
> records inserted:
> > 0
> > Tue Dec 18 13:44:35 UTC 2007: apel-pbs-log-parser - ------
> Processing
> > finished ------
> >
> > So, from what I understatnd:
> > The CE has all the insformation but this information acnnot
> be propagated
> > to the MON box. Our firewall is down for both the nodes(for
> local IP's)
> > and the rgma-gin is running on the mon box.
> >
> > Can you see something bad here?
> > Where should I search?
>
> /var/log/apel.log on the CE
>
> But first check on the MON that you have 'inspectTables' and
> _not_ 'inspectTable' in
> /opt/glite/etc/glite-apel-publisher/publisher-config-yaim.xml
>
> >
> >
> > thanks,
> > On Dec 18, 2007 4:26 PM, Kostas Koumantaros <
> [log in to unmask] <mailto:[log in to unmask]>> wrote:
> >
> > > Hi Asterios,
> > >
> > > the prolbem is that you site for some reason can not
> communicate with
> > > the R-GMA registry
> > > this either means that the registry is down or something
> prohibits
> > > your mon box to communicate with it.
> > >
> > > Cheers.
> > >
> > > K.
> > >
> > > On 18 Δεκ 2007, at 4:19 ΜΜ, Asterios Katsifodimos wrote:
> > >
> > > > Hello again,
> > > >
> > > > On our mon I can see this:
> > > >
> > > > ...
> > > > Thu Jul 27 02:47:32 UTC 2006: apel-publisher -
> Optimising table:
> > > > LcgRecords
> > > > Thu Jul 27 02:47:33 UTC 2006: apel-publisher - ****
> Combining
> > > > tables and republishing in LcgRecords ****
> > > > Thu Jul 27 02:47:33 UTC 2006: apel-publisher - Checking
> valid CPU
> > > > spec data exists
> > > > Thu Jul 27 02:47:33 UTC 2006: apel-publisher - CPU spec
> values found
> > > > Thu Jul 27 02:47:41 UTC 2006: apel-publisher - program
> aborted
> > > > org.glite.apel.core.ApelException:
> org.glite.rgma.RGMAException:
> > > > Unable to locate an available Registry Service
> > > > at org.glite.apel.publisher.AccountPublisher.<init>
> > > > (AccountPublisher.java:115)
> > > > at org.glite.apel.publisher.AccountManager.run
> > > > ( AccountManager.java:93)
> > > > at
> org.glite.apel.publisher.ApelPublisher.runJoinProcessor
> > > > (ApelPublisher.java:112)
> > > > at org.glite.apel.publisher.ApelPublisher.run
> > > > (ApelPublisher.java :68)
> > > > at org.glite.apel.publisher.ApelPublisher.main
> > > > (ApelPublisher.java:234)
> > > > Caused by: org.glite.rgma.RGMAException: Unable to
> locate an
> > > > available Registry Service
> > > > at org.edg.info.XMLSAXConverter.createRGMAException
> > > > (XMLSAXConverter.java:417)
> > > > at org.edg.info.XMLSAXConverter.createRGMAException
> > > > (XMLSAXConverter.java:427)
> > > > at org.edg.info.XMLSAXConverter.endElement
> > > > (XMLSAXConverter.java:370)
> > > > at org.apache.crimson.parser.Parser2.maybeElement
> > > > (Parser2.java:1720)
> > > > at
> org.apache.crimson.parser.Parser2.content(Parser2.java:
> > > > 1963)
> > > > at org.apache.crimson.parser.Parser2.maybeElement
> > > > (Parser2.java:1691)
> > > > at org.apache.crimson.parser.Parser2.parseInternal
> > > > (Parser2.java:667)
> > > > at org.apache.crimson.parser.Parser2.parse
> (Parser2.java:337)
> > > > at org.apache.crimson.parser.XMLReaderImpl.parse
> > > > (XMLReaderImpl.java:448)
> > > > at org.edg.info.XMLSAXConverter.convertXMLResponse
> > > > ( XMLSAXConverter.java:225)
> > > > at org.edg.info.ServletConnection.sendCommand
> > > > (ServletConnection.java:461)
> > > > at
> > > >
> org.glite.rgma.stubs.servlet.ProducerFactoryServletImpl.createInstance
>
> > > > (ProducerFactoryServletImpl.java :165)
> > > > at
> > > >
> org.glite.rgma.stubs.servlet.ProducerFactoryServletImpl.createPrimaryP
> > > > roducer(ProducerFactoryServletImpl.java :76)
> > > > at org.glite.apel.publisher.AccountPublisher.<init>
> > > > (AccountPublisher.java :108)
> > > > ... 4 more
> > > > Fri Jul 28 02:47:05 UTC 2006: apel-publisher - Read-in
> > > > configuration: [logenabled, j] [DBUsername=accounting,
> > > > DBURL=jdbc:mysql://mon101.grid.u
> > > > cy.ac.cy:3306/accounting
> <http://cy.ac.cy:3306/accounting> , DBPassword=****,
> site=CY-01-KIMON,
> > > > republish=missing]
> > > > Fri Jul 28 02:47:05 UTC 2006: apel-publisher - ------
> Starting the
> > > > apel application ------
> > > > Fri Jul 28 02:47:12 UTC 2006: apel-publisher -
> Optimising table:
> > > > EventRecords
> > > > Fri Jul 28 02:47:14 UTC 2006: apel-publisher -
> Optimising table:
> > > > GkRecords
> > > > ...
> > > >
> > > > What could be the problem?
> > > >
> > > > thanks again,
> > > >
> > > > On Dec 18, 2007 4:13 PM, Asterios Katsifodimos <
> > > > [log in to unmask] <mailto:[log in to unmask]> >
> wrote:
> > > > Hello *,
> > > >
> > > > We have a problem with our APEL accounting publisher.
> > > > It stopped publishing data for a month.
> > > >
> > > > However, in the log files on the CE I can see no
> suspicious things.
> > > >
> > > > Could you provide me with a checklist so I can start
> searching for
> > > > the problem?
> > > > I am very new to APEL.
> > > >
> > > >
> > > > thanks a lot in advance,
> > > > --
> > > > Asterios
> > > > CY-01-KIMON
> > > > University of Cyprus
> > > >
> > > >
> > > >
> > > > --
> > > > Asterios
> > > > _______________________________________________
> > > > Egee-sa1-tech mailing list
> > > > [log in to unmask] <mailto:[log in to unmask]>
> > > > https://mailman2.grnet.gr/mailman/listinfo/egee-sa1-tech
> > >
> > > Koumantaros Kostas, MSc
> > > Software Engineer / Grid Technologies
> > >
> > > -------------------------------------------------
> > > **Greek Research and Technology Network (GRNET)**
> > > Mesogion Avenue 56, 4th Floor, Room 4.1.6
> > > GR-11527, Ampelokipi, Athens, Greece
> > > -------------------------------------------------
> > >
> > > Tel.:+30 210 7474246
> > > Mob.: +30 697 7606622
> > > Fax.: +30 210 7474490
> > > Skype: kkoumantaros
> > > Email:[log in to unmask] <mailto:Email:[log in to unmask]>
> > > WWW: http://www.grnet.gr
> > >
> > >
> > >
> > >
> >
> >
> > --
> > Asterios
>
> > _______________________________________________
> > Egee-sa1-tech mailing list
> > [log in to unmask] <mailto:[log in to unmask]>
> > https://mailman2.grnet.gr/mailman/listinfo/egee-sa1-tech
> <https://mailman2.grnet.gr/mailman/listinfo/egee-sa1-tech>
>
> --
> Kyriakos Ginis
> Software Engineering Laboratory
> National Technical University of Athens
> _______________________________________________
> Egee-sa1-tech mailing list
> [log in to unmask] <mailto:[log in to unmask]>
> https://mailman2.grnet.gr/mailman/listinfo/egee-sa1-tech
> <https://mailman2.grnet.gr/mailman/listinfo/egee-sa1-tech>
>
>
>
>
> --
> Asterios
>
>
>
>
> --
> Asterios
--
Pablo Rey Mayo
Centro de Supercomputacion de Galicia
Avda. de Vigo. s/n (Campus Sur)
15706 Santiago de Compostela (Spain)
Tel: +34 981 56 98 10 ; Fax: +34 981 59 46 16
email: [log in to unmask] ; http://www.cesga.es/
------------------------------------------------
NOTA: Este mensaje ha sido redactado intencionadamente sin utilizar
acentos ni caracteres especiales, para que pueda ser visualizado
correctamente desde cualquier cliente de correo y sistema.
|
