The aim of this page is tracking the progressive decommissioning of SL5 from sites.
UMD is getting ready to support CentOS7. Supporting a new platform will require to decommission a previously support O.S. to make available new resources. EGI Operations will like to decommission Scientific Linux 5.
Please start tracking which sites are still using SL5 services: how many services, and for each service if still needed on SL5, if upgrades on SL5 services are expected). Also interesting to understand who is using Debian.
From September 2015 on, UMD3 will support only security fixes for SL5. Security updates will be available until March 2016.
All SL5 services must be decommissioned by April 2016. On February probes will start warning about SL5 services to be decommissioned.
Description of the process
- Procedure is https://wiki.egi.eu/wiki/PROC16_Decommissioning_of_unsupported_software
- Decommissioning deadline: April 30, 2016
- Decommissioning start date: January 31, 2016
- On Jan31 a new probe is deployed into the MW SAM and starts returning WARNING with SL5
- On Feb28 the probe starts returning CRITICAL.
- From Feb28 on, RODs follow up the service migration by creating operations ticket through Operations Dashboard (see https://wiki.egi.eu/wiki/PROC16_Decommissioning_of_unsupported_software#Escalation_phase)
[2016-03-01] Request to turn WARNINGs into CRITICAL: https://ggus.eu/?mode=ticket_info&ticket_id=118901#update#21
[2016-01-15] Preparation of probe requested https://ggus.eu/?mode=ticket_info&ticket_id=118901
[2016-01-15] UMD4 released with CentOS7
[2015-12-17] SL5 retirement announced at OMB https://indico.egi.eu/indico/getFile.py/access?contribId=5&resId=0&materialId=slides&confId=2383
[2015-07-13] SAM-nagios is still SL5, no SL6 available. ARGO will get fully centralized, there is no need to use SAM-nagios anymore, so regional instances won't be used anymore.
Services that are not probed by midmon
- dCache HINT: for installations exposing the web admin interface, you can go to: http://<hostname>:2288/info and check within the XML for the "<os>" tag
2016-04-18 Overall status
- 6 Top-BDII
- 30 Site-BDII
- 6 MyProxy
- 21 WMS and 11 LB
- 8 VOMS
- 2 ARGUS
- 27 CREAM-CE
- 2 QCG Computing
- 6 STORM
Reports by NGIs
Please report here which SL5 services are running, at which sites, and the corresponding dismission plan.
Only the regional SAM-Nagios box. This will be decommissioned as soon as the central monitoring is reliable.
The only SL5 is the one holding ngi.SAM (ngi-ro-nagios.nipne.ro).
Only NGI instances of SAM-nagios are on SL5.
Several sites: we’d like to clarify what timescale would be appropriate if we wanted to decommission SL5 pool/head nodes over the coming months rather than migrate.
With a note that squids are not covered in the probes/exceptions: are these included (some sites have SL5 squids)?
- BDII: SL5 BDII due for retirement
- CREAM - BDII: CentOS 7 being investigated
- CREAM: 1 VMHost, 2 x CREAM. CREAM to be moved to HTCondor CE (end of May)
- WMS/LB: Due for decommissioning shortly
- BDII: SL5 BDII due for upgrade after Easter
- SAM/VO Nagios SL5 only as with others.
Castor SRM systems. Small number of internal service machines. Database systems on Red Hat Enterprise 5. Timeline: SRMs and others - still to be defined.