I'm using a Netgear WNR834B router since yesterday.
It has an internal firewall and now I can't communicate with following projects:
Einstein@home
lhcathome
QMC@Home
PrimeGrid
Only SETI@home works fine. I can upload work to all project servers but scheduler requests always fail:
25.02.2007 00:00:02|Einstein@Home|Scheduler request failed: failed sending data to the peer
25.02.2007 00:00:02|Einstein@Home|Deferring communication for 1 min 0 sec
25.02.2007 00:00:02|Einstein@Home|Reason: scheduler request failed
25.02.2007 00:00:03||Access to reference site succeeded - project servers may be temporarily down.
25.02.2007 00:00:07|PrimeGrid|Sending scheduler request: Requested by user
25.02.2007 00:00:07|PrimeGrid|Reporting 9 tasks
25.02.2007 00:02:46||Project communication failed: attempting access to reference site
25.02.2007 00:02:47||Access to reference site succeeded - project servers may be temporarily down.
25.02.2007 00:02:48|PrimeGrid|Scheduler request failed: failed sending data to the peer
25.02.2007 00:02:48|PrimeGrid|Deferring communication for 32 min 26 sec
25.02.2007 00:02:48|PrimeGrid|Reason: scheduler request failed
25.02.2007 00:02:53|Einstein@Home|Sending scheduler request: Requested by user
25.02.2007 00:02:53|Einstein@Home|(not requesting new work or reporting completed tasks)
25.02.2007 00:05:39||Project communication failed: attempting access to reference site
25.02.2007 00:05:40||Access to reference site succeeded - project servers may be temporarily down.
25.02.2007 00:05:40|Einstein@Home|Scheduler request failed: failed sending data to the peer
25.02.2007 00:05:40|Einstein@Home|Deferring communication for 1 min 0 sec
25.02.2007 00:05:40|Einstein@Home|Reason: scheduler request failed
Any solution? I tried to deactivate the firewall, but it still did not work.
I think I have to allow access over port 80 for non-http, but it simply doesn't work with that stupid router...

---

Have you tried one of the schedulers in a web browser? Do you get an error message there too?

BOINC needs only HTTP for that btw., other ports are only needed for GUI and cmdline control.

Does it create sched_reply*.xml files on each contact and - if so - what are their contents?
If they look as expected, it can be a timing problem, I had that when a project server was overloaded.

Have you restarted the computer after installing the router? If BOINC sits in the memory, it might have missed the change in the gateway (otoh. ... it contacts Google so it is not very likely that this is the reason).

---

��u�

Have you tried one of the schedulers in a web browser? Do you get an error message there too?

Incomplete request received. But that's OK...

Does it create sched_reply*.xml files on each contact and - if so - what are their contents?
If they look as expected, it can be a timing problem, I had that when a project server was overloaded.

The sched_reply*.xml files are created, but they're empty. I can contact all projects when I connect without the router and there's no message that servers are overloaded.

Here another extract from my stdoutdae.txt where you can see PrimeGrid and LHC with different errors, BAM! with error and a successful SETI contact.
25.02.2007 15:01:23|PrimeGrid|Sending scheduler request: Requested by user
25.02.2007 15:01:23|PrimeGrid|Reporting 34 tasks
25.02.2007 15:03:29||Contacting account manager at http://bam.boincstats.com/
25.02.2007 15:04:09|PrimeGrid|Scheduler request failed: failed sending data to the peer
25.02.2007 15:04:09|PrimeGrid|Deferring communication for 1 min 0 sec
25.02.2007 15:04:09|PrimeGrid|Reason: scheduler request failed
25.02.2007 15:04:35|lhcathome|Sending scheduler request: Requested by user
25.02.2007 15:04:35|lhcathome|(not requesting new work or reporting completed tasks)
25.02.2007 15:06:14||Account manager error: http error
25.02.2007 15:06:14||Project communication failed: attempting access to reference site
25.02.2007 15:06:15||Access to reference site succeeded - project servers may be temporarily down.
25.02.2007 15:07:13||Project communication failed: attempting access to reference site
25.02.2007 15:07:14||Access to reference site succeeded - project servers may be temporarily down.
25.02.2007 15:07:14|lhcathome|Scheduler request failed: server returned nothing (no headers, no data)
25.02.2007 15:07:14|lhcathome|Deferring communication for 1 min 0 sec
25.02.2007 15:07:14|lhcathome|Reason: scheduler request failed
25.02.2007 15:07:20|PrimeGrid|Sending scheduler request: Requested by user
25.02.2007 15:07:20|PrimeGrid|Reporting 34 tasks
25.02.2007 15:10:04||Project communication failed: attempting access to reference site
25.02.2007 15:10:06||Access to reference site succeeded - project servers may be temporarily down.
25.02.2007 15:10:07|PrimeGrid|Scheduler request failed: failed sending data to the peer
25.02.2007 15:10:07|PrimeGrid|Deferring communication for 1 min 0 sec
25.02.2007 15:10:07|PrimeGrid|Reason: scheduler request failed
25.02.2007 15:12:59|SETI@home|Sending scheduler request: Requested by user
25.02.2007 15:12:59|SETI@home|(not requesting new work or reporting completed tasks)
25.02.2007 15:13:04|SETI@home|Scheduler RPC succeeded [server version 509]
25.02.2007 15:13:04|SETI@home|Deferring communication for 11 sec
25.02.2007 15:13:04|SETI@home|Reason: requested by project

I will try a clean deinstallation when I'm out of work next time. Or does anybody have another idea?

I will try a clean deinstallation when I'm out of work next time. Or does anybody have another idea?

Tried clean uninstallation with BOINC folder deletion, but it doesn't work. Only improvement is that I can contact BAM via the BOINC Manager now.

27.02.2007 15:01:45||Starting BOINC client version 5.8.15 for windows_intelx86
27.02.2007 15:01:45||log flags: task, file_xfer, sched_ops
27.02.2007 15:01:45||Libraries: libcurl/7.16.0 OpenSSL/0.9.8a zlib/1.2.3
27.02.2007 15:01:45||Data directory: C:\\Programme\\BOINC
27.02.2007 15:01:46||Processor: 1 GenuineIntel Mobile Intel(R) Celeron(R) CPU 1.50GHz [x86 Family 15 Model 2 Stepping 4] [fpu tsc sse mmx]
27.02.2007 15:01:46||Memory: 223.48 MB physical, 538.66 MB virtual
27.02.2007 15:01:46||Disk: 18.62 GB total, 16.09 GB free
27.02.2007 15:01:46||No general preferences found - using BOINC defaults
27.02.2007 15:01:47||This computer is not attached to any projects
27.02.2007 15:01:47||Visit http://boinc.berkeley.edu for instructions
27.02.2007 15:02:58||Contacting account manager at http://bam.boincstats.com/
27.02.2007 15:03:18||Account manager error: can't resolve hostname
27.02.2007 15:03:18||Project communication failed: attempting access to reference site
27.02.2007 15:03:20||Access to reference site failed - check network connection or proxy configuration.
27.02.2007 15:04:22||Contacting account manager at http://bam.boincstats.com/
27.02.2007 15:04:22||Account manager: BAM Host-ID: 20485
27.02.2007 15:04:22||Account manager contact succeeded
27.02.2007 15:04:23||Attaching to http://lhcathome.cern.ch/
27.02.2007 15:04:23||Attaching to http://www.primegrid.com/
27.02.2007 15:04:23||Attaching to http://spin.fh-bielefeld.de/
27.02.2007 15:04:23||Attaching to http://abcathome.com/
27.02.2007 15:04:24|http://lhcathome.cern.ch/|Master file download succeeded
27.02.2007 15:04:24|http://lhcathome.cern.ch/|Sending scheduler request: Project initialization
27.02.2007 15:04:24|http://lhcathome.cern.ch/|Requesting 1 seconds of new work
27.02.2007 15:07:03||Project communication failed: attempting access to reference site
27.02.2007 15:07:03|http://lhcathome.cern.ch/|Scheduler request failed: server returned nothing (no headers, no data)
27.02.2007 15:07:03|http://lhcathome.cern.ch/|Deferring communication for 1 min 0 sec
27.02.2007 15:07:03|http://lhcathome.cern.ch/|Reason: scheduler request failed
27.02.2007 15:07:06||Access to reference site succeeded - project servers may be temporarily down.
27.02.2007 15:07:07|http://www.primegrid.com/|Master file download succeeded
27.02.2007 15:07:07|http://www.primegrid.com/|Sending scheduler request: Project initialization
27.02.2007 15:07:07|http://www.primegrid.com/|Requesting 1 seconds of new work

---

Do you have a firewall on your own PC? Usually router firewalls are pretty basic and don't interfere with outgoing port 80 traffic.

---

Do you have a firewall on your own PC? Usually router firewalls are pretty basic and don't interfere with outgoing port 80 traffic.

No. Windows 2000 has no internal firewall and I'm not using any third party software firewall.
The crazy thing is, that I can't contact the project servers (except of SETI), even if I place the PC in a DMZ. So it might be a problem with the router's firmware (but it's the up-to-date version), causing the firewall to be enabled although PC is in DMZ, or it's not a firewall problem. But what could it be else? I can reach every corner of the internet with IE via the router, but to connect to the BOINC project servers I need to install a direct connection.

---

There are quite a few settings in a router. E.g. to get Skype to work i had to set up port forwarding to a specific machine. Here's a website explaining this and that on routers and what to set for many brands: http://www.portforward.com/

Port forwarding is for incoming connections, BOINC only needs outgoing port 80. Incoming 31416 useful if you want to control your client from outside, but that's all.

No. Windows 2000 has no internal firewall and I'm not using any third party software firewall.
The crazy thing is, that I can't contact the project servers (except of SETI), even if I place the PC in a DMZ. So it might be a problem with the router's firmware (but it's the up-to-date version), causing the firewall to be enabled although PC is in DMZ, or it's not a firewall problem. But what could it be else? I can reach every corner of the internet with IE via the router, but to connect to the BOINC project servers I need to install a direct connection.

DMZ is unlikely to help here, but likely to cause security problems on your computer. Stay away from it.

Let's not forget port 443 please, though few projects use it.

Do you know, which projects use that port?
As mentioned before, I can contact SETI, SETI-Beta, and the account manager BAM!, while ABC, LHC, QMC, Einstein, PrimeGrid, Predictor, Spinhenge are unreachable.

---

WCG is the only I know that uses HTTPS, so using port 443.

Hashbreaker uses the non-standard port 8700.

Remember this is all outgoing.

Remember this is all outgoing.

Are there any ports needed for incoming connections? I spoke to an expert from Netgear today at Cebit, and he could only imagine that BOINC needed some ports forwarded.

---

BOINC doesn't take *anything* incoming. Servers never send stuff to the clients, the clients ask for it. Then the server sends it back, in the same way a webserver sends you a website or image back to you when you request it.

The only exception is port 31416 for the BOINC Manager to connect to the client. It needs to be forwarded only if you need to administrate your clients remotely (from outside the firewall). I doubt you would do such thing...

I don't know if you can believe it, but I could not at the first sight...
I made a firmware update (manually, because the router didn't find the new version automatically), and it works now.
Thanks for your support, although I finally solved the problem on my own.

---