Changes between Initial Version and Version 1 of tier-one-meeting-20181120


Ignore:
Timestamp:
11/20/18 14:52:51 (21 months ago)
Author:
paul
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • tier-one-meeting-20181120

    v1 v1  
     1[[TOC(depth=0)]] 
     2= dCache Tier I meeting November 20, 2018 = 
     3[part of a [wiki:developers-meetings series of meetings]] 
     4 
     5== Present == 
     6 
     7dCache.org(), IN2P3(), Sara(), Triumf(), BNL(), NDGF(Jens), PIC(), KIT(Xavier), Fermi(), CERN() 
     8 
     9= Agenda = 
     10 
     11(see box on the other side) 
     12 
     13= Site reports = 
     14 
     15== KIT == 
     16 
     17Looks rather fine. 
     18 
     19Final downtime for updating dCache for LHCb to dCache v3.2. 
     20 
     21New problem: 
     22 
     23Reported on Wednesday via RT ticket 9545. 
     24 
     25Linked to the httpd domain runs out of memory -- the pool manager domain crashed with out-of-memory.  Does restarts default limit for heap space, rather than the configured limit. 
     26 
     27=== httpd domain === 
     28 
     29httpd domain still needs restarting (every 25--26 hours). 
     30 
     31This is described in RT Ticket 9537. 
     32 
     33Paul to clarify who is investigating this issue. 
     34 
     35=== ssh interface === 
     36 
     37Dmitry looking into ssh ticket, but hasn't done so. 
     38 
     39Paul to talk with Dmitry to remind him he promised to update the ticket. 
     40 
     41=== Lost 'well-known' status === 
     42 
     43Described in RT ticket 9542. 
     44 
     45A couple of pools lost their 'well-known' status in the dCache network. 
     46 
     47It seems to be related 
     48 
     49One of the ZK server dropped out of the  
     50 
     51One of the pool manager crashed in ATLAS instance. 
     52 
     53Xavier will update ticket with information about networking problem for ZK node, resulting ATLAS pool-manager crash, etc. 
     54 
     55== NDGF == 
     56 
     57Things are looking pretty good. 
     58 
     59Continued investigation for a number of issue. 
     60 
     61Two physical machines -- swapped the VMs on the physical machines and the anomalously high memory usage problem went away.  Both VMs now using the same amount of memory. 
     62 
     63NDGF will probably upgrade to the latest dCache v4.2 tomorrow. 
     64 
     65= Support tickets for discussion = 
     66 
     67[Items are added here automagically] 
     68 
     69= DTNM = 
     70 
     71Proposed: same time, next week.