Difference between revisions of "Tier1 Operations Report 2015-10-21"

Revision as of 10:57, 21 October 2015

Review of Issues during the week 14th to 21st October 2015.

We have been seeing problems with very high load on the AtlasTape Castor instance. This was leading to a big backlog of files waiting to go to tape as well as a big delay in files being brought online. Additinal disk servers have been added to this instance (the number of servers doubled from five to ten). At the time of the meeting Atlas have also throttled back the write requests.
LHCb have reported an ongoing low but persistent rate of failure when copying the results of batch jobs to other sites. They have also reported a current problem that sometimes occurs when writing these files to our Castor storage.
ILC reported that a CVMFS repository (/cvmfs/ilc.desy.de) was not available on the worker nodes. This was traced to a missing public key. This has now been fixed across the worker nodes.

Resolved Disk Server Issues

GDSS746 (AtlasDataDisk - D1T0) failed on Saturday (17th Oct). Following checks it was returned to service yesterday afternoon (Tuesday 20th). It was initially put into a readonly mode (overnight) as the RAID array verification ran. It was set in full production this morning. Three files that were truncated as the server failed have been declared lost to Atlas.

Current operational status and issues

The intermittent, low-level, load-related packet loss seen over external connections is still being tracked. Likewise we have been working to understand some remaining low level of packet loss seen within a part of our Tier1 network.
Long-standing CMS issues. The two items that remain are CMS Xroot (AAA) redirection and file open times. Work is ongoing into the Xroot redirection with a new server having been added in recent weeks. File open times using Xroot remain slow but this is a less significant problem.

Ongoing Disk Server Issues

GDSS637 (AtlasTape - D0T1) was taken out of service on Monday (19th Oct) as it was showing a disk drive /controller problem. There were no files waiting to go to tape. The problem is being investigated.

Notable Changes made since the last meeting.

A change has been made to the FTS3 service (at developers request) to fix a problem that caused some Atlas transfers to stall.
Yesterday (Tuesday 20th) four disk servers in tape-backed service classes were updated to SL6. This was a transparent operation.

Declared in the GOC DB

Advanced warning for other interventions

The following items are being discussed and are still to be formally scheduled and announced.

Upgrade of remaining Castor disk servers (those in tape-backed service classes) to SL6. This will be transparent to users.
Some detailed internal network re-configurations to enable the removal of the old 'core' switch from our network. This includes changing the way the UKLIGHT router connects into the Tier1 network.

Listing by category:

Databases:
- Switch LFC/3D to new Database Infrastructure.
Castor:
- Update SRMs to new version (includes updating to SL6).
- Update disk servers to SL6 (ongoing)
- Update to Castor version 2.1.15.
Networking:
- - Complete changes needed to remove the old core switch from the Tier1 network.
- Make routing changes to allow the removal of the UKLight Router.
Fabric
- Firmware updates on remaining EMC disk arrays (Castor, LFC)

Entries in GOC DB starting since the last report.

Open GGUS Tickets (Snapshot during morning of meeting)


GGUS ID	Level	Urgency	State	Creation	Last Update	VO	Subject
116866	Green	Less Urgent	On Hold	2015-10-12	2015-10-19	SNO+	snoplus support at RAL-LCG2 (pilot role)
116864	Green	Urgent	In Progress	2015-10-12	2015-10-14	CMS	T1_UK_RAL AAA opening and reading test failing again...

Availability Report

Key: Atlas HC = Atlas HammerCloud (Queue ANALY_RAL_SL6, Template 508); CMS HC = CMS HammerCloud


Day	OPS	Alice	Atlas	CMS	LHCb	Atlas HC	CMS HC	Comment
14/10/15	100	100	100	100	100	91	n/a
15/10/15	100	100	98	100	100	85	100	Single SRM test failure (ould not open connection to srm-atlas.gridpp.rl.ac.uk:8443)
16/10/15	100	100	100	98	100	89	100	Short problem with glexec in the early hours of the morning.
17/10/15	100	100	100	100	100	95	100
18/10/15	100	100	100	100	100	92	n/a
19/10/15	100	100	100	100	100	100	100
20/10/15	100	100	100	100	100	93	100

@@ Line 22: / Line 22: @@
 | style="background-color: #f8d6a9; border-bottom: 1px solid silver; text-align: center; font-size: 1em; font-weight: bold; margin-top: 0; margin-bottom: 0; padding-top: 0.1em; padding-bottom: 0.1em;" | Resolved Disk Server Issues
 |}
-* GDSS657 (LHCb_RAW - D0T1) was returned to service on Thursday (8th Oct). This server had failed on Saturday (3rd October). After being checked it was returned to service the following day in read-only mode. However, subsequent investigations to fix a problem on this server uncovered a further problem. As reported last week four files were found to be corrupt (partially written) from the time when the server first failed and these were reported to LHCb.
-* GDSS707 (AtlasDataDisk - D1T0) failed on Friday (16th Oct). This server had only been put back into service at the end of September. The server was drained of files over the weekend and the problem is being investigated.
 * GDSS746 (AtlasDataDisk - D1T0) failed on Saturday (17th Oct). Following checks it was returned to service yesterday afternoon (Tuesday 20th). It was initially put into a readonly mode (overnight) as the RAID array verification ran. It was set in full production this morning. Three files that were truncated as the server failed have been declared lost to Atlas.
 <!-- ***********End Resolved Disk Server Issues*********** ----->