Alert.png The wiki is deprecated and due to be decommissioned by the end of September 2022.
The content is being migrated to other supports, new updates will be ignored and lost.
If needed you can get in touch with EGI SDIS team using operations @ egi.eu.

Difference between revisions of "EGI-InSPIRE:SA1.5-QR8"

From EGIWiki
Jump to navigation Jump to search
(Copy of QR7 - not finished yet.)
 
(Repository text for Q8)
Line 2: Line 2:
= 1. Task Meetings =
= 1. Task Meetings =
<!--
<!--
Notes. Report here all task-specific meetings held. This includes (a) face-to-face meetings and (b) phone meetings. Make sure that for all task meetings participants are ALWAYS recorded either on indico from the registrants’ list, or in the minutes.  
Notes. Report here all task-specific meetings held. This includes (a) face-to-face meetings and (b)  
OMB meeting will be reported under task TSA1.1 only. Monday Operations meetings need to be reported under task TSA1.3 only. Training events will be recorded in the training event registry and need not be mentioned here.
phone meetings. Make sure that for all task meetings participants are ALWAYS recorded either on  
indico from the registrants’ list, or in the minutes. OMB meeting will be reported under task TSA1.1 only. Monday Operations meetings need to be reported  
 
under task TSA1.3 only. Training events will be recorded in the training event registry and need not  
 
be mentioned here.
-->
-->
{|  border="1" cellspacing="0" cellpadding="5" align="center"
{|  border="1" cellspacing="0" cellpadding="5" align="center"
Line 11: Line 16:
! style="width: 50%" | Outcome
! style="width: 50%" | Outcome
|-
|-
|29/3/12
|
|
|
|Accounting Workshop at CF12
|
|
|
|-
|-
Line 25: Line 30:
= 2. Main Achievements =  
= 2. Main Achievements =  
<!--
<!--
Note. This is a detailed account of progress over the previous quarter of activities within  the  task.  
Note. This is a detailed account of progress over the previous quarter of activities within  the   
 
task.  
-->
-->
Repository: Ran the production repository with no internal problems this quarter. There was one scheduled firewall outage of 45 minutes at RAL and a few very small network breaks which prevented the service receiving new data. This data would all have been received the next time the affected clients tried to publish. Total availability 99.79%.  
Repository: Ran the production repository with no internal problems this quarter. There were a few very small network breaks which prevented the service receiving new data. This data would all have been received the next time the affected clients tried to publish. The migration from SL4 to SL5 went smoothly for the old and the SSM-based servers.


Heavier support load than usual. Mainly due to rollout of CREAM CE and configuration errors during setup. It would be good to have second line support for APEL as for the other middleware. The APEL Team should only be providing third line support.
The support load was lower than Q7 as the rollout of CREAM CE has plateaued. It would be good to have second line support for APEL in DMSU as for the other middleware. The APEL Team should only be providing third line support.


The planned production release of the new infrastructure slipped beyond the end of this quarter.  
The production release of the new infrastructure happened in this quarter but none of the other
infrastructures have changed to use it yet. CERN is the most advanced, carrying out final tests
against the production instance.  


There is a test repository running all the time to receive tests from other sites developing their software against SSM the new STOMP and Python-based messaging layer which runs on the production EGI Messaging Infrastructure. All of the other existing and new accounting services have tested using SSM except SGAS where the developer has moved on and has not yet been replaced.
The test repository continues to run all the time to receive tests from other sites developing their software against SSM the new STOMP and Python-based messaging layer which runs on the production EGI Messaging Infrastructure. All of the other existing and new accounting services have done some testing using SSM, now including SGAS and GridSafe.  


Portal: No operating problems, the VM was configured with 3Gb (1Gb more) to avoid memory shortages on user decryption due to the ever growing number of user records. The new ActiveMQ connector should be a definitive solution to this problem.  The last release with the new codebase caused some early problems and regressions that were quickly fixed. After that, the number of user problems decreased sharply, and most tickets and mails are feature requests.
The test infrastructure successfully received ptotptye usage records from the resource providers in the EGI Cloud Task Force demonstration at EGI CF12. The strawman VM UR and the first script to harvest data from OpenNebula were done by the RAL team.


Since the Portal is a VM machine, it can grow easily to accommodate more load, but currently the growth in data size seems to be reasonable.
Portal:


= 3. Issues and Mitigation =
= 3. Issues and Mitigation =
Line 47: Line 56:
!scope="col"| Mitigation Description
!scope="col"| Mitigation Description
|-
|-
|SL4-5 migration requires all APEL servers to be down together for several days.
|
|Downtime announced well in advance. Primed TPM to respond to sites who post tickets having missed the broadcast.
|
|-
|-
|
|
Line 56: Line 65:
= 4. Plans for the next period =
= 4. Plans for the next period =
<!-- provide your text below -->
<!-- provide your text below -->
Start a production service receiving summaries from other accounting services over SSM and joining them with the old summary system. Once all systems have migrated the old database will be migrated to a new one and the old clients piped into that. This second step is likely to be in Q8. After the first step we will be ready to receive prototype records from storage and cloud iunfrastructures.
Draw up migration plans with the other infrastructures to move their summary publishing from the database insert to SSM. Once all systems have migrated the old database will be migrated to a new one and the old clients piped into that. This second step is likely to be in Q9, possibly Q10.


Portal:
Portal:
Update the operating system on production to SL5, and to SL6 on some future date.

Revision as of 17:46, 9 May 2012

1. Task Meetings

Date (dd/mm/yyyy) Url Indico Agenda Title Outcome
29/3/12 Accounting Workshop at CF12

2. Main Achievements

Repository: Ran the production repository with no internal problems this quarter. There were a few very small network breaks which prevented the service receiving new data. This data would all have been received the next time the affected clients tried to publish. The migration from SL4 to SL5 went smoothly for the old and the SSM-based servers.

The support load was lower than Q7 as the rollout of CREAM CE has plateaued. It would be good to have second line support for APEL in DMSU as for the other middleware. The APEL Team should only be providing third line support.

The production release of the new infrastructure happened in this quarter but none of the other infrastructures have changed to use it yet. CERN is the most advanced, carrying out final tests against the production instance.

The test repository continues to run all the time to receive tests from other sites developing their software against SSM the new STOMP and Python-based messaging layer which runs on the production EGI Messaging Infrastructure. All of the other existing and new accounting services have done some testing using SSM, now including SGAS and GridSafe.

The test infrastructure successfully received ptotptye usage records from the resource providers in the EGI Cloud Task Force demonstration at EGI CF12. The strawman VM UR and the first script to harvest data from OpenNebula were done by the RAL team.

Portal:

3. Issues and Mitigation

Issue Description Mitigation Description

4. Plans for the next period

Draw up migration plans with the other infrastructures to move their summary publishing from the database insert to SSM. Once all systems have migrated the old database will be migrated to a new one and the old clients piped into that. This second step is likely to be in Q9, possibly Q10.

Portal: