The past several times, only a few days. I might have died on a Friday evening and it was resolved by Monday. So this is truly an extended outage from the ones in the past. That is why I was wondering if he went on vacation or something.

Thank you for that info. I will hold out hope.

Did notice a change when this showed up in notices:

Enigma@Home: Notice from server
Enigma 0.76b needs 3633355906759967822163392472118519509897936333762380538799349237146650150447872889552536326477638749902077901780791787612123406287017170888293484786718292115578013593460801536.00 MB RAM but only 22116.19 MB is available for use.

... first time to see that one. Apparently enigma now wants more RAM than exists in the entire galaxy.

I see that message when there is no work. That message is nothing new.

[/quote]

I see that message when there is no work. That message is nothing new.
[/quote]

Thank you for letting me know about that.

What I would really like is to know if anyone here can put that number into words... I even tried a numbers to words conversion script, but it gives up about 30 digits before this number does. :-)

---

Yes. In words that is "more than six".

I see that message when there is no work. That message is nothing new.

Thank you for letting me know about that.

What I would really like is to know if anyone here can put that number into words... I even tried a numbers to words conversion script, but it gives up about 30 digits before this number does. :-)

According to http://www.isthe.com/cgi-bin/chongo/number.cgi, it is:
three septenquinquagintillion,
six hundred thirty three sexquinquagintillion,
three hundred fifty five quinquinquagintillion,
nine hundred six quattuorquinquagintillion,
seven hundred fifty nine trequinquagintillion,
nine hundred sixty seven duoquinquagintillion,
eight hundred twenty two unquinquagintillion,
one hundred sixty three quinquagintillion,
three hundred ninety two novemquadragintillion,
four hundred seventy two octoquadragintillion,
one hundred eighteen septenquadragintillion,
five hundred nineteen sexquadragintillion,
five hundred nine quinquadragintillion,
eight hundred ninety seven quattuorquadragintillion,
nine hundred thirty six trequadragintillion,
three hundred thirty three duoquadragintillion,
seven hundred sixty two unquadragintillion,
three hundred eighty quadragintillion,
five hundred thirty eight novemtrigintillion,
seven hundred ninety nine octotrigintillion,
three hundred forty nine septentrigintillion,
two hundred thirty seven sextrigintillion,
one hundred forty six quintrigintillion,
six hundred fifty quattuortrigintillion,
one hundred fifty tretrigintillion,
four hundred forty seven duotrigintillion,
eight hundred seventy two untrigintillion,
eight hundred eighty nine trigintillion,
five hundred fifty two novemvigintillion,
five hundred thirty six octovigintillion,
three hundred twenty six septenvigintillion,
four hundred seventy seven sexvigintillion,
six hundred thirty eight quinvigintillion,
seven hundred forty nine quattuorvigintillion,
nine hundred two trevigintillion,
seventy seven duovigintillion,
nine hundred one unvigintillion,
seven hundred eighty vigintillion,
seven hundred ninety one novemdecillion,
seven hundred eighty seven octodecillion,
six hundred twelve septendecillion,
one hundred twenty three sexdecillion,
four hundred six quindecillion,
two hundred eighty seven quattuordecillion,
seventeen tredecillion,
one hundred seventy duodecillion,
eight hundred eighty eight undecillion,
two hundred ninety three decillion,
four hundred eighty four nonillion,
seven hundred eighty six octillion,
seven hundred eighteen septillion,
two hundred ninety two sextillion,
one hundred fifteen quintillion,
five hundred seventy eight quadrillion,
thirteen trillion,
five hundred ninety three billion,
four hundred sixty million,
eight hundred one thousand,
five hundred thirty six
megabytes.

In short it is about 3.6 septenquinquagintillion MB, or about 3.6 novemquinquagintillion bytes.

---

I've contacted the registrar of the domain and the server hosting company asking them if they could get in touch with him. I haven't heard anything back as of yet.

I've contacted the registrar of the domain and the server hosting company asking them if they could get in touch with him. I haven't heard anything back as of yet.

Thank you.

I just hope he is okey.

Who knows, they might have the same contact info as we do. It can't hurt though. If I had to take a guess, he mentioned in the past that the server hosting is not the database and that he runs that at home. So the issue could be that, the database is down, which would explain why there is no work but the server status page shows everything as good.

On the server status page there is:
Risingson
Ravager

So two servers. One of which is 66.221.0.198 which is the hosted server which I believe is Risingson based upon this:

Since the early days the BOINC server is hidden behind another server, which acts like a proxy and load balancer. It handles some stuff (static files) on it's own and the rest is forwarded to the BOINC server via two VPNs.

So either Ravager is also hosted (we don't know) or it is at his place. With the server status page saying that everything is fine, there might be a third server that is the issue.

The main reason is that the server has only 2GB of RAM and it's working with huge databases. All returned results are stored, old results are copied to archive but only after the batch is set as inactive. Results from running batches are frequently accessed by the server daemons and they have to stay in the 'live' database.

I have to keep the BOINC part of database clean and as small as possible to avoid stressing the hardware too much. Even though the server was upgraded recently, the entire database still runs from a single drive.

Maybe the drive failed or some other database problem seems to be the likely cause. In the past there were no WU's as the hosting bill wasn't paid. Since we don't know the setup it could be anything.

So the issue could be that, the database is down, which would explain why there is no work but the server status page shows everything as good.

But the server status doesn't suggest everything is good, far from it.

5) number of active workunits with 0 results in progress. This should stay low, as it indicates work distribution problems.
But that number is just as high as 4), so there are work distribution problems.

All services are running though, so the server status we can see, everything is fine. That is why I'm thinking it is the database server. That is not shown on the server status page.

The number isn't 0:

Results ready to send6 2
High priority results7 0
Results sent last hour 0
Results in progress 29
Workunits waiting for validation 0
Workunits waiting for assimilation 10
Workunits waiting for deletion 0
Results waiting for deletion 0
Transitioner backlog (hours) 0

So there are 29 WU's still left. New work is not being generated though but the work generation process shows that it is running, but would require the database though. Once again, the database is not listed on the server status page.

waiting for new wu.

---

All services are running though, so the server status we can see, everything is fine. That is why I'm thinking it is the database server. That is not shown on the server status page.

You should not only look at the green part on the upper left of the screen, you should read the numbers on the right too, and read what they mean at the bottom.

there are work distribution problems.

All services are running though, so the server status we can see, everything is fine. That is why I'm thinking it is the database server. That is not shown on the server status page.

You should not only look at the green part on the upper left of the screen, you should read the numbers on the right too, and read what they mean at the bottom.

there are work distribution problems.

I know what they mean. There have been many projects that take a hiatus and they have the same numbers! usually it is because new work is being worked on to create more WU's. So while there is no new work, it is not because of the server listed on the server status page. It is a server behind the scenes which is in this is most likely the database server.

We have seen outages here many times and when they is no work, at least 50% of the time it was a server at the front-end as not all services were running or there was a high load on the server. That is not the case this time.

"6) - currently workunit generator wakes up when the result ready to send goes below 1k"

The workunit generator service is running, it just cannot generate more work. This would lead me to believe it is a database issue. The database status is not listed, so we don't know what the status of it is. Go read old posts and you will see that the last upgrade was back in 2013 and that it runs on a single drive. If that drive fails, what do you think happens? We also haven't heard from TJM, so we don't know if anything has been running on auto-pilot for awhile and the database has grown to large and is now having issues. There is care and feeding that is required and without it, eventually it will crash.

Maybe he is on vacation. This is what Poland has at a national level:
20 working days per year during the first 10 years of employment and 26 working days thereafter. Then there are 13 paid holidays as well. So it is not uncommon to see Europeans go on vacation for 3 to 5 weeks. I worked for a financial company and that is exactly what out European counterparts would do, leave for around a month. At times the entire office would be gone at the same time.

I received a response from the server hosting company, this is their current reply:

Hello,

Please stand by while we process this request.
We will update you once we have reached out to this Client.

Thank you very much for your patience in regards to this matter.
--

Maybe he is on vacation. This is what Poland has at a national level:
20 working days per year during the first 10 years of employment and 26 working days thereafter. Then there are 13 paid holidays as well. So it is not uncommon to see Europeans go on vacation for 3 to 5 weeks. I worked for a financial company and that is exactly what out European counterparts would do, leave for around a month. At times the entire office would be gone at the same time.

I lived my whole live (57 years till now) in Europe and I was never allowed to go on holidays without transferring my work to another co-worker. It is absolutely unthinkable to have the entire office gone at the same time, at least here in the Netherlands.

Still no word from him:

Hello,

We have received a request and are still attempting to reach out to you on behalf of the users of this
distributed computing project, and the server that is running it is having an
issue currently stated in the forum threads linked to us in the original message
:

G'day,
no work received since 18/3/17

regards,

The workunit generator service is running, it just cannot generate more work. This would lead me to believe it is a database issue.

Database is running, this whole site needs it for operating. Without DB you would not be able to login, use this forum, check stats, etc. - almost all things here depends on data in DB. However there is possibility that table(s) which store data about work units may be corrupted, so new WUs cannot be added.

---

Nothings says it is the same db used for everything. I see 15 valid WU's listed under my account and they are from August. I mentioned it to TJM in the past and nothing was done about it. So I have a feeling the database has grown to be too big and can no no longer work as desired.

In the past it was mentioned that the database was not hosted as it would cost too much money. So that tells me the WU aspect is housed at his place. It is also all on a single drive. I highly doubt that we are accessing that computer to post messages. So that means multiple databases.

Nothings says it is the same db used for everything. I see 15 valid WU's listed under my account and they are from August. I mentioned it to TJM in the past and nothing was done about it. So I have a feeling the database has grown to be too big and can no no longer work as desired.

In the past it was mentioned that the database was not hosted as it would cost too much money. So that tells me the WU aspect is housed at his place. It is also all on a single drive. I highly doubt that we are accessing that computer to post messages. So that means multiple databases.

I found this nice image which shows BOINC architecture. Every blocks on server server side may be hosted on the same or different physical machine. I hope this will help you understand how pieces of Enigma project may be divided between its hosts. Enigma is a bit different because it has two Web Serves, so on picture you would have another "Proxy/Load Balancer" block inserted between Data Server/Scheduling Server/Web Interfaces and Client Side. This still allows to have one DB. Additionally DB can be configured to allow access from multiple hosts, what can be used here if blocks are assigned to physical hosts in different way.

---

That diagram doesn't include project application component specific DB