Well, last of mine beamed up.

Wanders off to see if there are projects for todays issue first.

---

And why does "analyzing the back end" of the results require the shutting down of incoming data?

Money. There isn't enough of it to analyse the data already gathered, and keep sending out more to add to what's already been done.
If you could come up with, say $1.25 million- per year -for the next 10 years, then they could keep this part of Seti going for another 10 years, and analyse the data as well.

Send a letter to Bill Gates!!!!!!

I'm positive he could fund you for a Nebula upgrade and keep SETI going for another 10 with his pocket change!

8-)

Is the server closet where the magic first started?

---

Historian for the Defunct Riviera Hotel and Casino, Former Classic Seti@home user for Team Art Bell. Greetings from the High Desert!

So is this the "farewell" thread?

I may not have a high "credit" score but that's because much of my CPU time is from the days when computers were much slower, and recently because I've been doing folding instead of seti. But I loved this project, started when I was just a teenager. Same reason why I have a ridiculous username. Only thing I did before this was the distributed.net SHA challenges back in the mid 90's when I was just a kid.

I always loved this project, I'm sad they've run out of work for us to do. More telescope time is needed, so we can keep listening... someone is out there, we just have to find them.

So is this the "farewell" thread?

It is one of several "farewell" threads.

Tom M

---

A proud member of the OFA (Old Farts Association).

Is the server closet where the magic first started?

This is where it all began:

Before the "Big Reorg" - May 12, 2002


A few years later, s@h was hosted on:

Another Look at the Servers - December 22, 2008


Since then, the project outgrew the data bandwidth available to the lab, and then also outgrew the server closet itself. The main servers are now physically in a datacentre somewhere 'down the hill'.

Anyone know of any more recent pictures online?


Keep searchin',
Martin

---

See new freedom: Mageia Linux
Take a look for yourself: Linux Format
The Future is what We all make IT (GPLv3)

At the time of the move to the CoLo centre a lot of folks asked for pictures and we were very politely informed that it was highly unlikely that any would be allowed - I think Richard was one of the more recent Setizens to visit, and from his description of the visit security is pretty tight (I'm not sure if they did an "all body cavities" search, but.....)

---

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

:(

At the time of the move to the CoLo centre a lot of folks asked for pictures and we were very politely informed that it was highly unlikely that any would be allowed - I think Richard was one of the more recent Setizens to visit, and from his description of the visit security is pretty tight (I'm not sure if they did an "all body cavities" search, but.....)

No, they didn't go that far. But I had to be introduced and signed up by somebody 'on the list' (Jeff Cobb was able to do that), wear a name badge, and be escorted. Eric let me take a camera in, on the strict understanding that I only took pictures of SETI's kit. This is one which I've posted before:


Eric working on an upgrade to muarae2 - a twin to the server which drives these web pages. We were adding a PCIe card holding 4 M.2 HDDs, an extra 2 TB of storage if I remember right. We pulled the server out of the rack this side of the open door: behind the door you can see the 'crash cart' plugged into an access point to provide remote control of the server we needed to power down before removal.


A better view of the 'crash cart' - I think Eric had decided he needed to connect directly by this point - remote access wasn't quite giving him what he needed.


Eric replacing a failed drive in one of the storage arrays.


The general 'mixed server' rack from which we removed muarae2 for the upgrade.


Primarily a storage rack, including 'amigos' - the user-funded data storage server intended to hold Parkes data (next to bottom). The upper part of the rack holds Breakthrough Listen's storage arrays.


And for old times' sake - the original closet in the Space Science Lab, where it all started - pictured on the same day, 19 July 2019.

i know you said they had issues with that server. do you know what issues specifically? it wasn't related to seeing all of the m.2 drives was it?

to use that PCIe card, the server must support PCIe bifurcation. and you have to set the slot it's in to x4x4x4x4 for it to work. i can't tell how old that server is. looks like a Dell or HP 1U system of some sort.

---

Seti@Home classic workunits: 29,492 CPU time: 134,419 hours

i know you said they had issues with that server. do you know what issues specifically?

I only know what happened on that particular day. The card had arrived in a plain brown carton - enterprise version - none of the fancy packaging or silly fripperies you get in the retail market. Like manuals or accessory cables.

When we got back to the lab, Eric couldn't see the new drive at all. Quick online search for a user manual, and it turned out the card needed a supplementary power cable - which we hadn't fitted. At that point, the trail goes cold, as far as I'm concerned.

I'd say it's pretty likely that the server doesn't even support PCIe bifurcation. That is likely to be the main issue with that 4x m.2 PCIe card even if Eric ever gets the power cable he needs. you need more modern hardware for that thing to work. that card, takes 4x m.2 drives, which each have 4 PCIe lanes, and splits them amongst the x16 slot.

through cross referencing pictures and features I can see in your pics, I'm 99% certain that the motherboard in muarae2 is the Supermicro H8DGU-F. this is a very old AMD Opteron board.

This means that either it is not a true twin of muarae1, or the SSP specs for muarae1 are wrong or out of date. muarae1 (and 4) is listed as a dual hex-core 3.07GHz Xeon. the only chip fitting that description that I can find is the Intel Xeon Xeon X5675, which is another old Intel Xeon, socket 1366 with tri-channel DDR3 memory. very very different hardware than your picture shows.

what tipped me off to this discrepancy was seeing the single screw holding down the CPU heatsink. I knew no Intel CPUs used that kind of mounting.

---

Seti@Home classic workunits: 29,492 CPU time: 134,419 hours

some further digging looks like the card Eric picked up is either this or a clone of this one: RIIPTOP Quad M.2 NVMe SSD PCIe x16 Adapter, ASM2824

so this particular card *may* end up working, since it employs a PCIe switch connecting your 4x M.2 drives to a single PCIe x8 link. hard to say how such an old board will react to such a device though. sometimes combinations like this just don't work for unknown reasons. if it does work, you'll be limited to PCIe 2.0 speeds though since the motherboard in muarae2 can't do PCIe 3.0, effectively halving the total bandwidth.

---

Seti@Home classic workunits: 29,492 CPU time: 134,419 hours

Good sleuthing! Thanks and thanks to Richard.

But hey! That's still a useful speedup.

I've done far too much adding disks and bits to expand old servers myself, too often far beyond their original design and spec. The thing is... It really is amazing what you can get away with when using Linux systems!


There does look to be a LOT of disks in there...

Cue the s@h obelisk: "My God, it's full of disks"...


Keep searchin'!
Martin

---

See new freedom: Mageia Linux
Take a look for yourself: Linux Format
The Future is what We all make IT (GPLv3)

Big thanks for all the photos! Hard to believe my first account (too bad I can't find it) was on all that first generation equipment. Nice to see the progression of the servers over the decades.

Here's hoping to another chance down the road if available!

---

Historian for the Defunct Riviera Hotel and Casino, Former Classic Seti@home user for Team Art Bell. Greetings from the High Desert!

This means that either it is not a true twin of muarae1, or the SSP specs for muarae1 are wrong or out of date. muarae1 (and 4) is listed as a dual hex-core 3.07GHz Xeon. the only chip fitting that description that I can find is the Intel Xeon Xeon X5675, which is another old Intel Xeon, socket 1366 with tri-channel DDR3 memory. very very different hardware than your picture shows.

OK, did some sleuthing of my own. Matt tells at least some of the story in his Technical News stories.

Message 1382796 (19 Jun 2013): Server-news-wise, we did acquire another donated box - a 3U monster that actually contains four motherboards, each with 2 hexa-core Xeon CPUs and 72GB of memory, and 3 SATA drives. Despite being in one box, they are four distinct machines: muarae1, muarae2, muarae3, and muarae4.

Message 1649365 (5 Mar 2015): One of our servers (the four-headed monster that is muarae{1,2,3,4}) developed a weird power issue - muarae2 seems completely dead. Fair enough, but when you try to power cycle it for some reason muarae4 power cycles as well.

Message 1788204 (17 May 2016): our web server muarae1 ... went bonkers around the start of the year. Well, we finally got a new muarae1 so centurion is back to being centurion - a dedicated splitter, a storage server, and potentially a science database clone and analysis machine. We also got a muarae2 server which is a back up (and eventual replacement) for the seti.berkeley.edu web server.

My 'twin' remark was a distant memory (without checking) that all four muarae servers arrived together. I'd not registered stages 2 and 3 in the saga. I don't think the server descriptions on the SSP have, either.

There was for a short time SETI@CAMRAS
https://www.camras.nl/en/use/public-camras-data/
https://www.camras.nl/en/use/seti-camras/
https://gitlab.camras.nl/marc/setiatcamras
https://charon.camras.nl/public/setiatcamras/
but it seems to have been dropped like a hot potato :/

I want to switch to something, but would like to keep looking for ET's.
Thx all!

There aren't any others looking for ET's :(

Einstein@Home is looking for pulsars and assorted gravity waves only its not looking for intelligent signals too like Astro Pulse is/was.

There are a number of biology/life/medicine projects out there.

But nothing that is looking for ET.

I would be glad to be wrong. Let us know if you find one!!!!!

Tom

---

Apr 3, 1999 - May 3, 2020

After 20 years processing, this is a sad occasion. Seti has been part of my family for all of that time.
It is so sad to see it it go .... and just upgraded my PC too !!!
Farewell dear friend .. hopefully back in a different form ..

---

Hey Richard, thanks for the pics! Brings back good memories of my trip down there. Just got back into town, so this afternoon I'll have to take a dive into the forum here and see exactly what happened with our beloved project. I get the general gist of it, but would like to get more details if they are around. And Eric, if you see this, thanks again for the lemons, they were excellent!

---

Hey Richard, thanks for the pics! Brings back good memories of my trip down there. Just got back into town, so this afternoon I'll have to take a dive into the forum here and see exactly what happened with our beloved project. I get the general gist of it, but would like to get more details if they are around. And Eric, if you see this, thanks again for the lemons, they were excellent!

Al,
I would leave at least one system idling on S@H. Just in case.

If you want to keep your cpus VERY busy you can try some Coronavid19 research at Rosetta@Home. Need more available HDD space in the BOINC settings and as much ram in the Settings as possible. Default is 8 hour tasks on any sized computer.

Tom

---

A proud member of the OFA (Old Farts Association).