Difference between revisions of "Tier1 Operations Report 2019-05-20"

From GridPP Wiki
Jump to: navigation, search
(Created page with "==RAL Tier1 Operations Report for 13th May 2019== __NOTOC__ ====== ====== <!-- ************************************************************* -----> <!-- ***********Start Rev...")
 
()
 
(7 intermediate revisions by one user not shown)
Line 1: Line 1:
==RAL Tier1 Operations Report for 13th May 2019==
+
==RAL Tier1 Operations Report for 20th May 2019==
  
 
__NOTOC__
 
__NOTOC__
Line 8: Line 8:
 
{| width="100%" cellspacing="0" cellpadding="0" style="background-color: #ffffff; border: 1px solid silver; border-collapse: collapse; width: 100%; margin: 0 0 1em 0;"
 
{| width="100%" cellspacing="0" cellpadding="0" style="background-color: #ffffff; border: 1px solid silver; border-collapse: collapse; width: 100%; margin: 0 0 1em 0;"
 
|-
 
|-
| style="background-color: #b7f1ce; 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;" | Review of Issues during the week 5th May 2019 to the 13th May 2019.
+
| style="background-color: #b7f1ce; 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;" | Review of Issues during the week 13th May 2019 to the 20th May 2019.
 
|}
 
|}
* We are seeing high outbound packet loss over IPv6.  Central networking intend to perform a firmware update to the border routers (provisionally scheduled for Wednesday 22nd May).
+
* PPD had an unpatched Jenkins server which had been compromised.  This identified on Friday 17th May.  It was not a Grid resource (we believe it was an old MICE server).
* On 7th May, LHCb finished their migration from Castor to Echo.
+
* We are seeing high outbound packet loss over IPv6.  Central networking intend to perform a firmware update on Wednesday(22/5/2019), morning.
* We have temporarily halted the setup of Tape for Dune while we bring the new Tape robot online.
+
* DUNE jobs are running again at Tier-1.  This was reported 10 days ago but we believe had been a problem for longer.  
* CMS job failures were a result of incorrectly configured jobs trying to access all their data from a Finnish Tier-2.
+
* LHCb Castor (Disk) is now read only.
 
<!-- ***********End Review of Issues during last week*********** ----->
 
<!-- ***********End Review of Issues during last week*********** ----->
 
<!-- *********************************************************** ----->
 
<!-- *********************************************************** ----->
Line 194: Line 194:
 
! Solution
 
! Solution
 
|-
 
|-
| style="background-color: lightgreen;" | 141108
+
| style="background-color: lightgreen;" | 141262
| dune
+
| lhcb
 
| in progress
 
| in progress
| top priority
+
| very urgent
| 10/05/2019
+
| 21/05/2019
| 15/05/2019
+
| 21/05/2019
| Workload Management
+
| File Access
| Problem submitting DUNE jobs to RAL CEs
+
| Users are getting [FATAL] Auth failed
| EGI
+
| WLCG
 
|  
 
|  
 
|-
 
|-
Line 214: Line 214:
 
| Files vanished from RAL tape?
 
| Files vanished from RAL tape?
 
| EGI
 
| EGI
|
 
|-
 
| style="background-color: red;" | 140773
 
| lhcb
 
| in progress
 
| top priority
 
| 18/04/2019
 
| 08/05/2019
 
| Storage Systems
 
| Removal of Echo unbearably slow
 
| WLCG
 
 
|  
 
|  
 
|-
 
|-
Line 285: Line 274:
 
! Solution
 
! Solution
 
|-
 
|-
| style="background-color: lightgreen;" | 141105
+
| 141108
 +
| dune
 +
| verified
 +
| top priority
 +
| 10/05/2019
 +
| 17/05/2019
 +
| Workload Management
 +
| Problem submitting DUNE jobs to RAL CEs
 +
| EGI
 +
| For the avoidance of doubt, DUNE uses the same queue names as LIGO (or CMS).  and as can be seen we are happily running DUNE jobs. 
 +
 
 +
 
 +
Total: LIGO_UK_RAL_arc_ce01 Total LIGO_frontend OSG_Flock_frontend gpfrontend01_frontend
 +
Status Running 45 0 0 45
 +
Idle 5 5 0 0
 +
Waiting 5 5 0 0
 +
Pending 0 0 0 0
 +
Stage in 0 0 0 0
 +
Stage out 0 0 0 0
 +
Unknown 0 0 0 0
 +
Held 0 0 0 0
 +
Requested Max glideins 792 10 0 782
 +
Idle 0 0 0 0
 +
Client Monitor Claimed 90 0 0 90
 +
User Run Here 90 0 0 90
 +
User Running 90 0 0 90
 +
Unmatched 6391 0 0 6391
 +
User idle 0 0 0 0
 +
Registered 0 0 0 0
 +
Info age 33427 71 0 33356
 +
Troubleshoot Diff(Status: Running - CM: Registered) 45 0 0 45
 +
Diff(Status: Idle - Requested: Idle) 5 5 0 0
 +
|-
 +
| 141105
 
| ops
 
| ops
| solved
+
| verified
 
| less urgent
 
| less urgent
 
| 10/05/2019
 
| 10/05/2019
| 14/05/2019
+
| 17/05/2019
 
| Operations
 
| Operations
 
| [Rod Dashboard] Issues detected at RAL-LCG2
 
| [Rod Dashboard] Issues detected at RAL-LCG2
Line 304: Line 326:
 
andrew mcnab
 
andrew mcnab
 
|-
 
|-
| 140932
+
| 140965
| enmr.eu
+
| cms
| solved
+
| closed
| less urgent
+
| urgent
| 30/04/2019
+
| 02/05/2019
| 08/05/2019
+
| 17/05/2019
| Other
+
| CMS_Data Transfers
| how to install cvmfs on worker nodes
+
| Datatransfers T2_AT_VIenna -> T1_UK_RAL_Buffer failing
| EGI
+
| WLCG
| Hi Enrico,
+
| Possible corruption on the file coming from Vienna may have made the file un-deleteable at RAL. The file was replaced with a new copy at Vienna and transfer in debug stream is now green.
 
+
I'm going to make the assumption that as "it works perfectly", I can mark this ticket as solved.
+
 
+
Best regards
+
 
+
Darren
+
 
|-
 
|-
 
| 140887
 
| 140887
Line 333: Line 349:
 
Closing this ticket.
 
Closing this ticket.
 
|-
 
|-
| 140758
+
| 140660
| lhcb
+
| cms
 
| closed
 
| closed
 
| urgent
 
| urgent
| 17/04/2019
+
| 09/04/2019
| 08/05/2019
+
| 16/05/2019
| File Access
+
| CMS_Central Workflows
| lhcbUser svcClass not working as it should ?
+
| FIle read issues for Workflows where data is located at T1_UK_RAL
 
| WLCG
 
| WLCG
| Hi guys,
+
| The gridmap  has been replaced and we appear to be working again.  As such I'm going to close this ticket as implementation of the the full solution is beyond the scope of this ticket.
 
+
I'm assuming I can now resolve this one again?
+
 
+
Cheers
+
 
+
D.
+
 
|}
 
|}
 
|}<!-- **********************End Availability Report************************** ----->
 
|}<!-- **********************End Availability Report************************** ----->
Line 374: Line 384:
 
! Comments
 
! Comments
 
|-
 
|-
| 2019-05-06
+
| 2019-05-14
 
| 100
 
| 100
 
| 100
 
| 100
Line 383: Line 393:
 
|  
 
|  
 
|-
 
|-
| 2019-05-07
+
| 2019-05-15
| 100
+
 
| 100
 
| 100
 
| 100
 
| 100
 +
| 99
 
| 100
 
| 100
 
| 100
 
| 100
Line 392: Line 402:
 
|  
 
|  
 
|-
 
|-
| 2019-05-08
+
| 2019-05-16
 
| 100
 
| 100
 
| 100
 
| 100
Line 401: Line 411:
 
|  
 
|  
 
|-
 
|-
| 2019-05-09
+
| 2019-05-17
 
| 100
 
| 100
 
| 100
 
| 100
Line 410: Line 420:
 
|  
 
|  
 
|-
 
|-
| 2019-05-10
+
| 2019-05-18
 
| 100
 
| 100
 
| 100
 
| 100
Line 419: Line 429:
 
|  
 
|  
 
|-
 
|-
| 2019-05-11
+
| 2019-05-19
 
| 100
 
| 100
 
| 100
 
| 100
 
| 100
 
| 100
 
| 100
 
| 100
| style="background-color: red;" | 74
 
| style="background-color: red;" | 74
 
|
 
|-
 
| 2019-05-12
 
 
| 100
 
| 100
 
| 100
 
| 100
| 100
 
| 100
 
| style="background-color: red;" | 24
 
| style="background-color: red;" | 24
 
 
|  
 
|  
 
|-
 
|-
| 2019-05-13
+
| 2019-05-20
 
| 100
 
| 100
 
| 100
 
| 100
Line 463: Line 464:
 
! Day !! Atlas HC !! CMS HC !! Comment
 
! Day !! Atlas HC !! CMS HC !! Comment
 
|-
 
|-
| 2019-05-06 || 100 || 99 ||
+
| 2019-05-14 || 100 || 99 ||  
|-
+
| 2019-05-07 || 100 || 99 ||  
+
 
|-
 
|-
| 2019-05-08 || 100 || 100 ||  
+
| 2019-05-15 || 100 || 99 ||  
 
|-
 
|-
| 2019-05-09 || 100 || 100 ||  
+
| 2019-05-16 || 100 || 100 ||  
 
|-
 
|-
| 2019-05-10 || 100 || 98 ||  
+
| 2019-05-17 || 100 || 100 ||  
 
|-
 
|-
| 2019-05-11 || 100 || 100 ||  
+
| 2019-05-18 || 100 || 99 ||  
 
|-
 
|-
| 2019-05-12 || 100 || 100 ||  
+
| 2019-05-19 || 100 || 100 ||  
 
|-
 
|-
| 2019-05-13 || 100 || 100 ||  
+
| 2019-05-20 || 100 || 100 ||  
 
|-
 
|-
 
|}  
 
|}  

Latest revision as of 09:03, 21 May 2019

RAL Tier1 Operations Report for 20th May 2019

Review of Issues during the week 13th May 2019 to the 20th May 2019.
  • PPD had an unpatched Jenkins server which had been compromised. This identified on Friday 17th May. It was not a Grid resource (we believe it was an old MICE server).
  • We are seeing high outbound packet loss over IPv6. Central networking intend to perform a firmware update on Wednesday(22/5/2019), morning.
  • DUNE jobs are running again at Tier-1. This was reported 10 days ago but we believe had been a problem for longer.
  • LHCb Castor (Disk) is now read only.
Current operational status and issues
Resolved Castor Disk Server Issues
Machine VO DiskPool dxtx Comments
- - - - -
Ongoing Castor Disk Server Issues
Machine VO DiskPool dxtx Comments
- - - - -
Limits on concurrent batch system jobs.
  • ALICE - 1000
Notable Changes made since the last meeting.
  • NTR
Entries in GOC DB starting since the last report.
Service ID Scheduled? Outage/At Risk Start End Duration Reason
- - - - - - - -
Declared in the GOC DB
Service ID Scheduled? Outage/At Risk Start End Duration Reason
- - - - - - - -
  • No ongoing downtime
Advanced warning for other interventions
The following items are being discussed and are still to be formally scheduled and announced.

Listing by category:

  • DNS servers will be rolled out within the Tier1 network.
Open

GGUS Tickets (Snapshot taken during morning of the meeting).

Request id Affected vo Status Priority Date of creation Last update Type of problem Subject Scope Solution
141262 lhcb in progress very urgent 21/05/2019 21/05/2019 File Access Users are getting [FATAL] Auth failed WLCG
140870 t2k.org in progress less urgent 25/04/2019 14/05/2019 Data Management - generic Files vanished from RAL tape? EGI
140447 dteam on hold less urgent 27/03/2019 14/05/2019 Network problem packet loss outbound from RAL-LCG2 over IPv6 EGI
140220 mice in progress less urgent 15/03/2019 15/05/2019 Other mice LFC to DFC transition EGI
139672 other in progress urgent 13/02/2019 30/04/2019 Middleware No LIGO pilots running at RAL EGI
GGUS Tickets Closed Last week
Request id Affected vo Status Priority Date of creation Last update Type of problem Subject Scope Solution
141108 dune verified top priority 10/05/2019 17/05/2019 Workload Management Problem submitting DUNE jobs to RAL CEs EGI For the avoidance of doubt, DUNE uses the same queue names as LIGO (or CMS). and as can be seen we are happily running DUNE jobs.


Total: LIGO_UK_RAL_arc_ce01 Total LIGO_frontend OSG_Flock_frontend gpfrontend01_frontend Status Running 45 0 0 45 Idle 5 5 0 0 Waiting 5 5 0 0 Pending 0 0 0 0 Stage in 0 0 0 0 Stage out 0 0 0 0 Unknown 0 0 0 0 Held 0 0 0 0 Requested Max glideins 792 10 0 782 Idle 0 0 0 0 Client Monitor Claimed 90 0 0 90 User Run Here 90 0 0 90 User Running 90 0 0 90 Unmatched 6391 0 0 6391 User idle 0 0 0 0 Registered 0 0 0 0 Info age 33427 71 0 33356 Troubleshoot Diff(Status: Running - CM: Registered) 45 0 0 45 Diff(Status: Idle - Requested: Idle) 5 5 0 0

141105 ops verified less urgent 10/05/2019 17/05/2019 Operations [Rod Dashboard] Issues detected at RAL-LCG2 EGI The problem has been solved.
Details about the solution ---------

Passing tests now, thanks!


andrew mcnab

140965 cms closed urgent 02/05/2019 17/05/2019 CMS_Data Transfers Datatransfers T2_AT_VIenna -> T1_UK_RAL_Buffer failing WLCG Possible corruption on the file coming from Vienna may have made the file un-deleteable at RAL. The file was replaced with a new copy at Vienna and transfer in debug stream is now green.
140887 atlas closed urgent 27/04/2019 13/05/2019 File Transfer UK RAL-LCG2 ransfer error with: srm-ifce err: Communication error on send WLCG This is not a RAL issue, but a problem with Wuppertalprod already ticketed at https://ggus.eu/index.php?mode=ticket_info&ticket_id=140883 .

Closing this ticket.

140660 cms closed urgent 09/04/2019 16/05/2019 CMS_Central Workflows FIle read issues for Workflows where data is located at T1_UK_RAL WLCG The gridmap has been replaced and we appear to be working again. As such I'm going to close this ticket as implementation of the the full solution is beyond the scope of this ticket.

Availability Report

Day Atlas Atlas-Echo CMS LHCB Alice OPS Comments
2019-05-14 100 100 100 100 100 100
2019-05-15 100 100 99 100 100 100
2019-05-16 100 100 100 100 100 100
2019-05-17 100 100 100 100 100 100
2019-05-18 100 100 100 100 100 100
2019-05-19 100 100 100 100 100 100
2019-05-20 100 100 100 100 100 100
Hammercloud Test Report
Target Availability for each site is 97.0% Red <90% Orange <97%
Day Atlas HC CMS HC Comment
2019-05-14 100 99
2019-05-15 100 99
2019-05-16 100 100
2019-05-17 100 100
2019-05-18 100 99
2019-05-19 100 100
2019-05-20 100 100

Key: Atlas HC = Atlas HammerCloud (Queue RAL-LCG2_UCORE, Template 841); CMS HC = CMS HammerCloud

Notes from Meeting.