The little "Bruno", have now more than 5,000,000 waiting "db purge" and of course it will increase more and more.
Very soon it will be a bottleneck on it ! It status is "not running". Strange, Not the server itself, because other programs seems run normally.
I only try to understand !
I am sure all the team at Berkely try and do the best they can. Since degin december there are more and more problems after weekly outage.

And no any news from the staff. I understand they prefer to solve the problems. But it will be nice if they publish informations. It will reduce the amount of message on the forume from people who not knows what happen.

Regards from Belgium.

To mederator's, not move to "Panic mode". It is quiet different. It is only about and to the staff. Nothig to make with "crunching" or "panic"...

---

And another little oddity:

17/01/2018 19:29:53 | SETI@home | Sending scheduler request: To report completed tasks.
17/01/2018 19:29:53 | SETI@home | Reporting 1 completed tasks
17/01/2018 19:29:53 | SETI@home | Not requesting tasks: "no new tasks" requested via Manager
17/01/2018 19:29:59 | SETI@home | Started download of arecibo_181.png
17/01/2018 19:29:59 | SETI@home | Started download of sah_40.png
17/01/2018 19:29:59 | SETI@home | Started download of sah_banner_290.png
17/01/2018 19:29:59 | SETI@home | Started download of sah_ss_290.png

This machine got one sporadic resend this morning, but apart from that, nothing at all has downloaded (from SETI) since maintenance started yesterday.

So the new server code gives us more pretty pictures?

modification to the .png? Would removing those and allowing download of new ones correct that?

---

Gents, kind regards from Belgium, but it is not my question !!!

---

OK....
Purging is not a continuous process, but runs sporadically.

---

Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?

On Tuesday, they loaded a new version of the BOINC software onto the servers. There was a bug in the new version of the db_purge script, introduced during unrelated changes elsewhere. It probably wasn't noticed at Beta, because Beta wants to keep previous results visible for comparison during testing, and normally doesn't purge them.

David fixed the bug, Eric checked it and approved it (slightly reluctantly, judging by his comment), and it seems to have done the trick:


I think the slight rise recently is normal: those will be the new tasks created yesterday, serving their normal 24 hour period of visibility before being eligible for purging.