LHC@home

since a few days ago, when looking up the tasks list of my various hosts, I noticed now and then a task with status "download error".
Now, one of my hosts would not let me download any task at all.
Whenever I push the update button, new tasks are showing up in the download manager, but a few seconds later the status says "download error".

The BOINC events log shows the following:

23.05.2024 14:50:05 | LHC@home | Started download of boinc_job_script.ZHGBc6
23.05.2024 14:50:05 | LHC@home | [error] File JX0KDmOONT5nsSi4apGgGQJmABFKDmABFKDm4ySLDmBVWKDmZoIKVo_input.tar.gz has wrong size: expected 479968, got 0
23.05.2024 14:50:05 | LHC@home | [error] Checksum or signature error for JX0KDmOONT5nsSi4apGgGQJmABFKDmABFKDm4ySLDmBVWKDmZoIKVo_input.tar.gz
23.05.2024 14:50:06 | LHC@home | Incomplete read of 520.000000 < 5KB for boinc_job_script.ZHGBc6 - truncating
23.05.2024 14:50:06 | LHC@home | Finished download of boinc_job_script.ZHGBc6
23.05.2024 14:50:06 | LHC@home | [error] File boinc_job_script.ZHGBc6 has wrong size: expected 17537, got 0
23.05.2024 14:50:06 | LHC@home | [error] Checksum or signature error for boinc_job_script.ZHGBc6

so, what's this now?

Edit: I am seeing this behaviour on some of my other hosts as well :-(

We have to wait, because of new Server upgrade yesterday.
Seeing the same and have stopped download.

what happens now is:
shortly after a new task shows up in the BOINC Manager, it is being "cancelled by project" before the download starts.

Hopefully these problems will be straightened out soon :-)

Canceling Tasks is in other projects also. Saw this in one project today.
With this function, you don't need to run it a third, or fourth time.
For me, Atlas running well atm.

There is something wrong with the server status page data. The 'In progress' value jumped from about 14000 to 54000 on 23rd of May at about 8:30 but the Jobs graph doesn't show any significant change for that time.
See here: https://grafana.kiska.pw/d/boinc/boinc?orgId=1&var-project=lhc@home&from=now-2d&to=now&refresh=30m So maybe the data on server is screwed.

---

by now I am experiencing the "download failure" problem on all of my hosts (all Windows).

does anyone of you people NOT have this problem?

No problems downloading here, but then I'm only doing one or two downloads at a time

Two absurd WUs here.
Look at this of my hosts:
https://lhcathome.cern.ch/lhcathome/results.php?hostid=10847676
Old WU's, done last week.
They are announced for downloading again and again. But no download at all.
The worst is, they block download of new Atlas WUs.

My other hosts are downloading like a charm.

ISP say no problems, BUT
Cern have 1 MBit/s for Atlas atm.
A few weeks ago was the same Problem.

Now the download problem seems to be different to what it was before - see BOINC event log:

24.05.2024 22:08:58 | LHC@home | Started download of Q6aLDmdJ5T5n7Olcko1bjSoqABFKDmABFKDmOtvXDmPHdKDmcz8DZn_EVNT.38776201._000085.pool.root.1
24.05.2024 22:08:58 | LHC@home | Started download of Q6aLDmdJ5T5n7Olcko1bjSoqABFKDmABFKDmOtvXDmPHdKDmcz8DZn_input.tar.gz
24.05.2024 22:08:58 | LHC@home | Started download of boinc_job_script.yNAYSM
24.05.2024 22:08:59 | LHC@home | Giving up on download of Q6aLDmdJ5T5n7Olcko1bjSoqABFKDmABFKDmOtvXDmPHdKDmcz8DZn_EVNT.38776201._000085.pool.root.1: permanent HTTP error
24.05.2024 22:08:59 | LHC@home | Finished download of Q6aLDmdJ5T5n7Olcko1bjSoqABFKDmABFKDmOtvXDmPHdKDmcz8DZn_input.tar.gz (0 bytes)
24.05.2024 22:08:59 | LHC@home | Finished download of boinc_job_script.yNAYSM (0 bytes)

Same here. I have started another boinc instance to get around the problem. Hope these tasks get sortet out asap.

One upload with 1,1 GByte atm?
Uploadspeed not more as 500 kBits. (30 minutes)

also, something is wrong with the tasks list in the user's webpage.

for this host, as can be seen, 5 ATLAS tasks are shown as being in process:
https://lhcathome.cern.ch/lhcathome/results.php?hostid=10815905

In reality, only 1 is in process, all others got finished day before yesterday and yesterday. So why are they not shown as finished and validated (or not validated)?

In the past days, there seem to be quite a number of problems over there at LHC&home

for this host, as can be seen, 5 ATLAS tasks are shown as being in process:
https://lhcathome.cern.ch/lhcathome/results.php?hostid=10815905

These prevent new WUs for this host!
It's impossible to abort them.

Do you have in Boincmanager -> Filetransfer for this Atlas-Tasks transfers which are open?
Had last Night also Atlas-Tasks with Problems, but are now finished and validated.

Do you have in Boincmanager -> Filetransfer for this Atlas-Tasks transfers which are open?

Nothing in Filetransfer.
As soon as I set "Allow new work" they appear for some seconds as downloads, but there is no transfer. It changes time of download, nothing else.

Edit:
A new try gave new WUs. But the 2 blocking tasks are still present.

@David Cameron
Please delete these 2 WU from server.

QlULDmVieS5n7Olcko1bjSoqABFKDmABFKDmOtvXDmvMaKDmojnWSn
XLgKDmdkhR5nsSi4apGgGQJmABFKDmABFKDm4ySLDm6FPKDmFCgdxm

They are misconfigured (end date in 2027).

Hostid 10847676 trys to download them again and again.
Can't interrupt this. Aborting does not help.

For example

Anwendung
ATLAS Simulation 3.01 (native_mt)
Name
QlULDmVieS5n7Olcko1bjSoqABFKDmABFKDmOtvXDmvMaKDmojnWSn
Status
Herunterladen fehlgeschlagen
erhalten
Mo 27 Mai 2024 13:29:44
Ablaufdatum
So 04 Jul 2027 09:54:10
Ressourcen
6 CPUs
Geschätzter Berechnungsaufwand
43’200 GFLOPs
Ausführbare Datei
wrapper_26015_x86_64-pc-linux-gnu

Has anyone actually had a successful task download for atlas since this issue started? Ive been running theory since it started but I try and give atlas a go once a day and all of my downloads continue to fail.

Has anyone actually had a successful task download for atlas since this issue started?

Yes.
Had a short power outage 2 days ago that caused all tasks to fail.
Hence, I restarted from scratch.
ATLAS was last (today).
So far without any download issues and some valids.


BTW:
@Saturn911
David left CERN more than a year ago.
https://lhcathome.cern.ch/lhcathome/forum_thread.php?id=5989&postid=48019

Two today:
BipKDmO9fV5n9Rq4apOajLDm4fhM0noT9bVo2ijZDmF6FKDmBB9tyn received at 10:31
https://lhcathome.cern.ch/lhcathome/result.php?resultid=411404266

zaBMDmcviV5n9Rq4apOajLDm4fhM0noT9bVo2ijZDmABGKDm8fILCn received at 14:51
https://lhcathome.cern.ch/lhcathome/result.php?resultid=411405886

Both are sitting in the queue ready to sart