I tried to beat the servers and resupply after my morning reschedule. But of course timed it most imperfectly ... as usual. Oh well, the CPU on Numbskull gets a breather today which matches up I guess with a cloudy day and low generation.

I did this to myself the last outage. Should have known better. It just bothers me to not run the most efficient BLC CPU tasks on my 1700X but for some reason the servers continue to send me almost exclusively Arecibo CPU tasks. That is why I reschedule.

An interesting idea of using the CPU just for SETI and use the GPU's for Einstein solely. On watts per credit basis, I do believe that Numbskull has shown me that the CPU tasks are beating the 970's. If I could somehow get the servers to exclusively send me only BLC tasks I wouldn't have to use the rescheduler on that machine. Then the GPU's move to Einstein would make a lot of sense.

But I have wanted to upgrade the GPU's on that machine to match the 1070's on all my other crunchers. Seems a disservice to hamstring the potential performance of Numbskull now that it has this new kickass CPU.

I'm getting an error message for every page:
Notice: Trying to get property of non-object in /disks/carolyn/b/home/boincadm/projects/sah/html/user/hosts_user.php on line 63
Is it just me...?

---

Not just you. Everybody gets that if trying to access any page other than the home page. It is what you'll get if trying to access your stats for example since the project is down for the normal Tuesday maintenance period.

9 hrs and still down.

I'm getting an error message for every page:
Notice: Trying to get property of non-object in /disks/carolyn/b/home/boincadm/projects/sah/html/user/hosts_user.php on line 63
Is it just me...?

Since they redid the web page schemas, you get this.
With the old version, the error was handled and you just got message telling you the site was down for maintenance.
This has been identified as an error, but apparently is not being addressed.

---

Now we are back, whoo hoo!

Jim,

I suggest that you raise it over at Main in number crunching.

Done, several times previously, in the thread on the new web development.

---

Anyone else here getting friend requests from someone through email, that never show up in your account? If you do, can you please post the account's name?

---

SETI boards are acting up... setiathome.berkeley.edu took too long to respond. Pages time out after about 60 seconds. Not sure if it's affecting retrieving / returning tasks yet.

Reserving my seat for this weeks Tuesday Matinee :-)

Ooops - getting a shed load of DNS errors when trying to get onto the website. Hope it's just a local problem and not a global one...

Been no problem here (so far).

No problem accessing the web site, but the aforesaid 'shedload of DNS errors' accessing the data servers - specifically upload and scheduler, but probably download as well.

There's a lot of router updating work scheduled by IST this month (IST Service Status), but our particular 208.68.240.* address range doesn't seem to be listed. With no results coming back in, they should notice quickly enough...

For all intents and purposes seti is off air at the present moment.

Suspend network activity and wait till after the weekly outrage , least people can still post so ain't exactly like the world's ending ( just yet , do wish people had more in their lives than " sittin in front of a computer posting on a obsure message board ;-) ) . Sun's out sky's blue birds are tweetling away , roll up a couple and put a few tinnies in me rucksack and relax down the river, to nice to be stuck indoors :-)

Richard - did you notice the date stamp on the second entry on the list of outages:

Scheduled Maintenance AirBears2, Eduroam, Calvisitor, ucbvpn

Outage Type: SCHEDULED MAINTENANCE
Date Submitted: Thursday, April 13, 2017
Outage Start/End Time: 0530 – 0600
Groups impacted: Campus
Equipment: airbears2, eduroam, calvisitor, ucbvpn

Has Berkeley found a time machine as there a few more announcements submitted in the future...

Open forum index, loads.
Open Number Crunching, and...

Project down for maintenance
Please check back in a few hours.

---

The outrage commences early this week....

Open forum index, loads.
Open Number Crunching, and...

Project down for maintenance
Please check back in a few hours.

I'm getting server not found.

---

Richard - did you notice the date stamp on the second entry on the list of outages:

Yes, but they were all scheduled maintenance/outages - planned in advance.

And yes, they do notify users in advance, for the problems they know about at least. But they don't seem to schedule work to start at 3 AM PDT, which is when our DNS went down (well, between 03:05 and 03:10 PDT, anyway).

At the time I posted, the website was working normally, with everything shown as running and up-to-date on the SSP - but now it's all showing as routine maintenance. I presume they can let that run as normal using fixed IP addresses internally, while they fight with the DNS machine for the rest of us.

Just in case we need them later (if the servers come back up before the DNS), here are - I think - the current server addresses.

208.68.240.118		setiboincdata.ssl.berkeley.edu	# upload server Oct 2016
# 208.68.240.126	setiboinc.ssl.berkeley.edu	# scheduler Oct 2016
# 208.68.240.127	boinc2.ssl.berkeley.edu		# download server Oct 2016

Upload is the only one we can use during maintenance - and it's accepting files just fine. Leave the other two commented out until/unless you need them.