Difference between revisions of "RAL Tier1 weekly operations castor 29/03/2019"
From GridPP Wiki
(→Achievements this week) |
(→Operation problems) |
||
Line 37: | Line 37: | ||
== Operation problems == | == Operation problems == | ||
+ | * gdss733 crashed and has been removed from production for two days. Now back into prod, no issues so far. | ||
+ | * TimF was doing a tape verify for Diamond (fd1866) | ||
+ | ** The verify was supposed to verify first 10, last 10, and random 10 files in the middle, but for some reason the test run on the all tape, therefore taking a lot of time more than the usual 30 minutes. Data has been inaccessible for a long time. | ||
+ | ** We would like the test not to run on the read-only Diamond tape server | ||
+ | ** Tim is looking at the reason why this happened | ||
* ATLAS are periodically submitting silly SAM tests that impact availability and cause pointless callouts. | * ATLAS are periodically submitting silly SAM tests that impact availability and cause pointless callouts. | ||
** Rob has created a ticket with Tim, work is being done. | ** Rob has created a ticket with Tim, work is being done. |
Revision as of 10:57, 29 March 2019
Contents
Standing agenda
1. Achievements this week
2. Problems encountered this week
3. What are we planning to do next week?
4. Long-term project updates (if not already covered)
5. Special topics
6. Actions
7. Review Fabric tasks
1. Link
8. AoTechnicalB
9. Availability for next week
10. On-Call
11. AoOtherB
Achievements this week
- Old facd0tl disk servers have been in read-only mode for one week, can be decommissioned next week
- Facilities headnodes requested on VMWare, ticket not done yet. Haven't heard anything from RIG yet.
- Willing to accept delays on this until ~May.
- Queued behind new disk, tape robot and a number of Diamond ICAT tasks.
- Acceptance testing of the new tape robot completed
- New-style tape server installation ongoing.
- Tape library for CASTOR-side testing in progress now
- Aquilon disk servers ready to go, also queued behind tape robot.
Operation problems
- gdss733 crashed and has been removed from production for two days. Now back into prod, no issues so far.
- TimF was doing a tape verify for Diamond (fd1866)
- The verify was supposed to verify first 10, last 10, and random 10 files in the middle, but for some reason the test run on the all tape, therefore taking a lot of time more than the usual 30 minutes. Data has been inaccessible for a long time.
- We would like the test not to run on the read-only Diamond tape server
- Tim is looking at the reason why this happened
- ATLAS are periodically submitting silly SAM tests that impact availability and cause pointless callouts.
- Rob has created a ticket with Tim, work is being done.
- CASTOR metric reporting for GridPP.
- Looking for clarity on precisely what metrics are relevant, and given CASTOR's changed role, what system RA should report on.
Plans for next few weeks
- Examine further standardisation of CASTOR pool settings.
- CASTOR team to generate a list of nonstandard settings and consider whether they are justified.
- CASTOR side tape robot testing.
Long-term projects
- New CASTOR WLCGTape instance.
- LHCb migration is with LHCb at the moment, they are not blocked. Mirroring of lhcbDst to Echo complete.
- CASTOR disk server migration to Aquilon.
- Change ready to implement.
- Deadline of end of April to get Facilities moved to generic VM headnodes and 2.1.17 tape servers.
- Ticket with Fabric team to make the VMs.
- RA working with James to sort out the gridmap-file distribution infrastructure and get a machine with a better name for this than castor-functional-test1
Actions
- AD wants us to make sure that experiments cannot write to that part of namespace that was used for d1t0 data: namespace cleanup/deletion of empty dirs.
- Some discussion about what exactly is required and how this can be actually implemented.
- CASTOR team proposal is to switch all of these directories to a fileclass with a requirement for a tape copy but no migration route; this will cause an error whenever any writes are attempted.
- RA to look at making all fileclasses have nbcopies >= 1.
- Problem with functional test node using a personal proxy which runs out some time in July.
- Rob met with Jens, requested an appropriate certificate.
Staffing
- RA out for the next two weeks, at HEPiX next week, on A/L the week after.
AoB
On Call
GP on call