If you set no venue at the project(s) you are attached to, or set the venue/computer location to '-', you get host location: none. The hostID is just a single entry in the state file (clientstate.xml) so there are quite easy ways to make a completely new install have the old hostID even if you have totally lost (hard disk destroyed) the old data directory. Content is available under GNU Free Documentation License 1.2 unless otherwise noted. I've double-checked everything and can't find the problem. The message 'Host location: none' has nothing to do with localhost or how BOINC Manager connects to BOINC. Just copy BOINC data folder from an old HDD to a new SSD That's the easiest way. This page was last edited on 19 March 2023, at 13:59. The BOINC server for this project is connected to several other key parts of the nanoHUB. -passwd passwd: Use passwd as password for RPC authentication. Where was, obviously, the name of the workunit file passed to make_work. The nanoHUBHome server is moving to a new physical location. In this way hosts are associated with countries. It has been suggested that its a firewall problem, but gufw doesnt have a profile for BOINC. BOINC users, during the registration process, can specify their country. Trying boinccmd always returns cant connect to local host. BOINC Manager is greyed out and shows connecting to localhost. BOINC has worked previously but now wont run. All seemed good, but when my client connected to the project, it downloaded everything except the actual file I passed to make_work. Have just upgraded online from 18.10 to 19.04 (desktop 64 bit). I set up my app folder, added the wrapper and the worker as described, set up my XML file and templates, and ran make_work. For the ssh connection, this is the user login. username is a string containing the username. It is quite robust and will report any errors allowing you to create alerts when that happens. For the http/https connection, this is the full path to the project page (omitting the server name). You can find it here: It features: - A python self-contained script that reads the BOINC clientstate.xml file and outputs statistics into the Prometheus. For the ssh connection, this is the directory where the BOINC project is located. I was using the instructions from the BOINC wiki as a guide dir is an RBOINC project directory on the BOINC server. However I am not able to get the app to process data on the clients. I successfully set up the project, and wrote my code using the "wrapper" setup. Once the BOINC software is installed in a machine, the server. I'd usually blame this on dodgy firewalls/av/routers etc, but these are clean XP installs with no junkware installed, and the project used to run fine until Sudden Death Syndrome hits it.Īny ideas? A thorough google just comes up with the usual stuff about firewalls, proxies and the like Oh, these are all BOINC 5.10.I am trying to set up a BOINC server to do some serious crunching. The BOINC framework consists of two layers which operate under the clientserver architecture. It can only connect to one client at a time. In the past I've seen BURP and Pirates die in the same way. It communicates with a running client using port 31416 over TCP/IP. The project used to run fine, but just suddenly diesįired up YAFU again recently, all of the hosts that I run it on connect and get work fine. I've checked all the local BOINC settings, no firewalls/antivirus/proxies in use (apart from router firewall), hosts file OK, can ping/tracert the project from the host etc etc. It looks as if the project is down, but it isn't as I can still see it in a browser on that host, and other hosts hanging off the same network switch still communicate without problems. Not sure where to ask this, so I'll try here as it's probably the best fitĮvery now and then, a random host (all XP32) loses the ability to contact a random BOINC project, it just gets "project temporarily unavailable".
0 Comments
Leave a Reply. |