ATLAS downloads extremely slow

Author	Message
computezrmle Volunteer moderator Volunteer developer Volunteer tester Help desk expert Send message Joined: 15 Jun 08 Posts: 2679 Credit: 286,757,479 RAC: 83,674	Message 42509 - Posted: 16 May 2020, 7:24:06 UTC 1 h to download an EVNT file. That's far too much. Just 1 example out of many: Sa 16 Mai 2020 08:16:29 CEST \| LHC@home \| Started download of 0jKMDmJGytwn9Rq4apoT9bVoABFKDmABFKDmhr8SDmABFKDmcDAHmm_EVNT.19651959._000500.pool.root.1 Sa 16 Mai 2020 09:15:21 CEST \| LHC@home \| Finished download of 0jKMDmJGytwn9Rq4apoT9bVoABFKDmABFKDmhr8SDmABFKDmcDAHmm_EVNT.19651959._000500.pool.root.1 ID: 42509 · Reply Quote

maeax Send message Joined: 2 May 07 Posts: 2276 Credit: 178,190,523 RAC: 115,378	Message 42511 - Posted: 16 May 2020, 8:21:22 UTC - in response to Message 42509. +1 ID: 42511 · Reply Quote

Crystal Pellet Volunteer moderator Volunteer tester Send message Joined: 14 Jan 10 Posts: 1461 Credit: 9,851,536 RAC: 3,668	Message 42513 - Posted: 16 May 2020, 8:43:44 UTC - in response to Message 42511. The same yesterday with a CMS-vdi file ID: 42513 · Reply Quote

djoser Send message Joined: 30 Aug 14 Posts: 145 Credit: 10,847,070 RAC: 0	Message 42524 - Posted: 17 May 2020, 12:38:02 UTC There seem to be a general network problem. Look at the timecodes from one of my Atlas native tasks: [2020-05-17 12:01:36] Probing /cvmfs/atlas.cern.ch... OK [2020-05-17 12:14:17] Probing /cvmfs/atlas-condb.cern.ch... OK Nearly 14 minutes no response... I also notice that current tasks seems to be interrupted while working and therefore take longer overall. Why mine when you can research? - GRIDCOIN - Real cryptocurrency without wasting hashes! https://gridcoin.us ID: 42524 · Reply Quote

Jim1348 Send message Joined: 15 Nov 14 Posts: 602 Credit: 24,371,321 RAC: 0	Message 42525 - Posted: 17 May 2020, 14:00:19 UTC Last modified: 17 May 2020, 14:06:41 UTC I am down to 7 out of 8 full cores on an i7-9700 for the past two hours, with "no tasks available". I don't know if that is related or not. I expect not; it is just the same old stuff. (I have only native ATLAS and CMS selected, and no other BOINC projects). EDIT: I just downloaded 9 more native ATLAS, to fill me up. They came in fast. I have a good connection to CERN, when they make use of it. ID: 42525 · Reply Quote

djoser Send message Joined: 30 Aug 14 Posts: 145 Credit: 10,847,070 RAC: 0	Message 42526 - Posted: 17 May 2020, 14:33:26 UTC Last modified: 17 May 2020, 14:34:38 UTC Here is another example from a different task: [2020-05-17 12:43:01] Checking for CVMFS [2020-05-17 12:43:06] Probing /cvmfs/atlas.cern.ch... OK [2020-05-17 12:43:15] Probing /cvmfs/atlas-condb.cern.ch... OK [2020-05-17 12:58:05] Probing /cvmfs/grid.cern.ch... OK 15 minutes for a response. Console command 'cvmfs_config probe' has the same problem. I just restartet my whole network and internet infrastructure, just in case it's ISP related. BTW: I don't have any trouble downloading new tasks either, they come as fast as usual. Why mine when you can research? - GRIDCOIN - Real cryptocurrency without wasting hashes! https://gridcoin.us ID: 42526 · Reply Quote

djoser Send message Joined: 30 Aug 14 Posts: 145 Credit: 10,847,070 RAC: 0	Message 42529 - Posted: 17 May 2020, 20:23:58 UTC Things are getting worse. This is a result of a task running on a machine that worked flawlessly for over 9 months now: https://lhcathome.cern.ch/lhcathome/result.php?resultid=273695262 On that same machine the cpu efficiency went down from 0.96 to 0.6. Why mine when you can research? - GRIDCOIN - Real cryptocurrency without wasting hashes! https://gridcoin.us ID: 42529 · Reply Quote

LHC@home