Message boards :
Number crunching :
Enhancement Request: Better info about FAQ-Updates
Message board moderation
| Author | Message |
|---|---|
 Yeti Send message Joined: 2 Sep 04 Posts: 453 Credit: 193,569,815 RAC: 11,186    |
Hello together, I had already crunched Theory, LHCb, CMS and was the last time focussing on Atlas. When I crunched the above projects I had the needed rules in my firewall opened as advised here:  As we can not subscribe to this page no one gets an info, when things change and we need additional rules / open Ports. Could you please find a solution where we can subscribe and get noticed of any changes ? Thank you very much  Supporting BOINC, a great concept ! |
Viking69 Send message Joined: 24 Jul 05 Posts: 56 Credit: 5,602,899 RAC: 9 
|
I would be leery for any application that 'needs' to have ports opened in a firewall. The whole reason for BOINC is for all the applications to share the resources and features for doing data crunching. If these VM's needed special privileges, then they are not BOINC work and should provide their own process for this. I personally had issues with the VM's causing my very robust system to lock up, forcing a hard power cycle and losing dozens if not hundreds of hours of work. LHC work is very important, but getting out of the BOINC structure is not what I signed up for a LONG time ago. Let's crunch for our future. |
Magic Quantum Mechanic Send message Joined: 24 Oct 04 Posts: 1127 Credit: 49,749,586 RAC: 10,234 
|
I wish I had some idea that another change was being made with these new huge .vdi's had to be d/l so I don't have 7 computers trying to each d/l 249.77MB X7 Then I could try doing that after midnight because nobody in the entire country has a DSL service that likes letting you do that during the day and love to throttle down the speed so slow that it takes 3 hours each to do this. I have had to do this over and over the last 6 years with these VB tasks. I just went to get more LHC tasks on one and first thing I see is the new .vdi and it started out at an ok speed but I am watching it slow down from over 100k to now under 70K and it most likely will soon be down to 30K On a good day 20mins is the fastest......on a typical day it will be 3 hours each. AND this does not help trying to start all the other tasks so they can get the connection with the Condor and credentials and HTCondor ping. Volunteer Mad Scientist For Life |
©2024 CERN