PRPNet 3.0.0 Released

[Xentar]

During the night, I got another exception, after a runtime of many hours.
Then I restarted the server at 8:12, and now (8:40) I see another exception.
The address of the error is always the same: 0x00446fcf, so it should be easy to find. Just the address, that is tried to read, is different.

Will check the log file, and try to debug this, when I am at home today evening.
Is VS Express 2008 ok, or do I need the professional version?

[rogue]

Originally Posted by Xentar

During the night, I got another exception, after a runtime of many hours.
Then I restarted the server at 8:12, and now (8:40) I see another exception.
The address of the error is always the same: 0x00446fcf, so it should be easy to find. Just the address, that is tried to read, is different.

Will check the log file, and try to debug this, when I am at home today evening.
Is VS Express 2008 ok, or do I need the professional version?

Visual Studio Express 2008 will work.

[rogue]

Originally Posted by rogue

Right now I know of five issues.

1) The garbled webpage.
2) "No available candidates on server" on Windows. (Lennart's issue)
3) Client times out waiting for a response from the server when connecting.
4) Client times out waiting for a response from the server when getting work. (Ian's issue)
5) A server crash.

I have an update on these for the next release

1) Fixed.
2) I don't have a export from Lennart yet, but I will do some of my own testing to see if I can reproduce the problem.
3 and 4) I have seen some odd behavior with sockets which I cannot explain. I made a couple of code changes when opening sockets, such as adding TCP_NODELAY. So far I haven't seen another issue, but that doesn't mean that these issues have been resolved. I won't know anything until I get this running on Max's server and through a bunch of clients at it.
5) Fixed. It was a problem of a printf missing an argument that the format string needed.

I'll add #6. Memory leaks in 3.3.x. I have reverted the GUI output code to the 3.2.x version and have re-implemented the changes from 3.3.x in a different way. I intend to test this out tonight. I'll know by tomorrow if the memory leak problem is fixed.

[mdettweiler]

Originally Posted by rogue

I'll add #6. Memory leaks in 3.3.x. I have reverted the GUI output code to the 3.2.x version and have re-implemented the changes from 3.3.x in a different way. I intend to test this out tonight. I'll know by tomorrow if the memory leak problem is fixed.

PRPnet has a GUI? I'm assuming you're referring to PFGW here, then?

[rogue]

Originally Posted by mdettweiler

PRPnet has a GUI? I'm assuming you're referring to PFGW here, then?

Duh! Yes. I was just about to correct that mistake in my post when I saw your reply.

[rogue]

Originally Posted by Xentar

I tried using PRPNet 3.2.0 for my sierp base19, server system is windows XP proffesional, SP 3. 6 Clients active.
After a few hours, I get an access violation "in 0x00446fcf, reading of memory address 0xeffd8e68".

Program continues execution until I click OK, but I am not sure, if data is lost?

BTW, I suspect this is the same server bug that I've seen, which will be fixed in the next release. If desired, I can send you the code change necessary to fix the bug.

[Xentar]

Originally Posted by rogue

BTW, I suspect this is the same server bug that I've seen, which will be fixed in the next release. If desired, I can send you the code change necessary to fix the bug.

When is the next planned release?
And, could you please try to explain the situation, in which is error occurs? when it happens while writing to the database, I think there will be data lost?

[rogue]

Originally Posted by Xentar

When is the next planned release?
And, could you please try to explain the situation, in which is error occurs? when it happens while writing to the database, I think there will be data lost?

As to your first question, unknown. I am traveling this week (for work) and won't have the time to dedicate to testing. I'm considering a couple of changes that I might spend time on while I'm away.

The error occurs when you find a prime or PRP and the thread to "unhide" it runs. That thread fails when trying to print a message to the log. No data is lost as a result of the bug.

[Xentar]

Thanks for the explanation.
That could be the reason, why it appears here so often: Sierp base 19, where I am running the server, has 970 k left. With 6 clients, I find about 5 primes per day I like this software

[rogue]

Originally Posted by Xentar

Thanks for the explanation.
That could be the reason, why it appears here so often: Sierp base 19, where I am running the server, has 970 k left. With 6 clients, I find about 5 primes per day I like this software

Thank you. I appreciate the compliment. It is unfortunate that it doesn't quite work so well for high throughput projects. NPLB has experienced a number of problems that are unique to that project. It is my goal to get there, but it is taking longer than I had anticipated.

[Xentar]

With the fix, the access violation didn't occurr again.
But today in the morning, there was another error:

[2010-02-17 04:43:08 GMT] ODBC Information: SQL_INVALID_HANDLE:
[2010-02-17 04:43:08 GMT] <MailAddress> (01): Test 1266381725 for candidate 325816*19^36314+1 was not found
[2010-02-17 04:43:45 GMT] Connect to database failed: [Microsoft][ODBC Driver Manager] Fehler beim SQLAllocHandle-Aufruf für Treiber auf SQL_HANDLE_ENV, native code=0
[2010-02-17 04:43:45 GMT] Not connected to database. Rejecting client connection.
[2010-02-17 04:43:45 GMT] Connect to database failed: [Microsoft][ODBC Driver Manager] Fehler beim SQLAllocHandle-Aufruf für Treiber auf SQL_HANDLE_ENV, native code=0
[2010-02-17 04:43:45 GMT] Not connected to database. Rejecting client connection.
[2010-02-17 04:45:01 GMT] Connect to database failed: [Microsoft][ODBC Driver Manager] Fehler beim SQLAllocHandle-Aufruf für Treiber auf SQL_HANDLE_ENV, native code=0
[2010-02-17 04:45:01 GMT] ODBC Information: SQL_INVALID_HANDLE:
[2010-02-17 04:48:05 GMT] Connect to database failed: [Microsoft][ODBC Driver Manager] Fehler beim SQLAllocHandle-Aufruf für Treiber auf SQL_HANDLE_ENV, native code=0
[2010-02-17 04:48:05 GMT] ODBC Information: SQL_INVALID_HANDLE:
....
and so on

The error message is in german, I can try to translate it:
[2010-02-17 04:43:45 GMT] Connect to database failed: [Microsoft][ODBC Driver Manager] Fehler beim SQLAllocHandle-Aufruf für Treiber auf SQL_HANDLE_ENV, native code=0
means something like
[2010-02-17 04:43:45 GMT] Connect to database failed: [Microsoft][ODBC Driver Manager] Error while calling SQLAllocHandle of driver at (or in?) SQL_HANDLE_ENV, native code=0

Seems, like the server lost the database connection, but I don't know why. I was sleeping at 4:40am