| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Retired Forums
Forum: Member-to-Member Support [Read Only]
Thread: UD Monitor 4.35 beta 3, now with WCG support |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 24
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
RM[stranger w/ 2 posts in 4 mos.]; I new that Sourceforge would be here [somewere since NOVEMBER!?!?!?]
For us 'NEWBIEZ, plz elaborate on what UDmonitor will do for the good of mankind [as we know it ]. " Until now it supported the grid.org and Folding@home clients only, but since the WCG Agent is not very different from the grid.org's one, it was quite easy to add support for it, too." What about GOOGle-labs F@H...? can I really leave my PC to crunch for 2 weeks while on vacation to do all 4 WU's?? sorry RT@MOT for 'Stepping on YOUR thread , But... ps: when the linux agent gets here, I want to see Techknow SourceForge ppl beat Techknow /. ! hehehehehhe |
|||
|
||||
|
ski1939
Senior Cruncher Freeport, Illinois, USA Joined: Nov 17, 2004 Post Count: 209 Status: Offline Project Badges: 
|
Thanks RT -
----------------------------------------I waited until the current wu was finished (about 8 hours) and removed the agent from start>programs>startup and then added ud monitor as you suggested and then rebooted. I think this solves the issue of having the monitor start automatically. However, I think I still have other problems because it wanted to reregister the device for the wu in cache slot 5 that the monitor wanted to start running. So I did and it gave me the msg "another instance of the grid agent is already running" again. Anyway, after registering another slot I tried to send/receive and it did not send the result in slot1 so I lost that 8 hour wu result. I don't know what I did wrong. I really wish it was easier to set up and use, but I guess its too complex for me. I don't want to loose anymore work units. Again, I want to thank you for your help RT, I think your suggestion works fine and my remaining problems has to do with something else I have done wrong. ski1939 (team USMC)  |
||
|
|
RT
Master Cruncher USA - Texas - DFW Joined: Dec 22, 2004 Post Count: 2636 Status: Offline Project Badges: 
|
If I remember correctly, the UDMon is supposed to manage all the slots for you. He is to always run the oldest WU and when it is finished, (after the agent sends the results and gets a new WU), he takes control from that agent and looks through all the slots for the oldest WU and starts up that agent. The process continues "forever".
----------------------------------------You are correct in that you should be able for force a send/receive at which time he is supposed to look at all the agents and find out if there are any ready to send... (That is 100% I think). If so, he is supposed to start that agent and let it run its little benchmark, send and the fetch a new WU. I have known times in the past that this has not worked...but then when he came to that slot again, it did send the WU and fetch another. So.... the work in your slot one may not be lost. Let him round-robin the thing before you give up on it. Now as to the agent already running. That is a new one on me. Either you somehow are getting one started outside the control of UDMon or UDMon is not functioning correctly supposedly he would be attempting to start more than one --- I have never seen or heard of this. Did you ever try to run UDMon as a service? If you did, an agent can be getting started that way. If you have not tried to run it as a service, and you cannot find another copy of the agent in your startup, then I suppose that it must be in the UDmon. I think I would contact the author.. who is located on the menu here I think I will stop here so that I don't tell you more than I know...if I haven't already  Very best of luck. |
||
|
|
ski1939
Senior Cruncher Freeport, Illinois, USA Joined: Nov 17, 2004 Post Count: 209 Status: Offline Project Badges:
|
Hi Robert (RT) - Thanks again for the support and encouragement -
----------------------------------------I've been thinking that somehow I did not install the monitor or cache slots properly, remember I was adding 5 cache slots and each time it requested that the slot/device get registered I got the msg "another instance of the grid agent is already running" - in reacting to it each time I exited the agent thinking that the devices had gotten registered. since all 5 slots did get loaded with a wu. However after the first WU finished under control of the monitor (slot 1) I was again asked to reregister the device, which I did and again got the message "another instance of the grid agent is already running". but the completed wu in slot 1 was not transmitted to the server - so I stopped running the monitor and now just run the agent. Today when I checked my stats on the server it only showed 2 devices my original and the one that I reregistered. So the other 4 slots/devices apparently did not get registered. The reregistered slot/device is now my current device name when I display preferences from the agent - not my original device name. I know a little about services/processes (display in task manager), but I have never set one up to run the agent or any thing else as I don't know how and I can't see where 2 instances of the agent were running when I got the msg, again looking at task manager displays. I did notice one thing the monitor version is dated 08DEC04 and there was a new rosetta version on 16DEC04 and a new version of the agent on 10DEC04, not that it has anything to do with this problem but I thought I would mention it. I'll take your suggestion and send Roman an email asking that he review this thread http://www.worldcommunitygrid.org/forums/wcg/viewthread?thread=478 and see what he can recommend. I hope I am not discouraging others from trying the monitor because it really is an excellent extension to the agent, especially the ability to download several wu and have them completed without returning to the server each time. This capability can maximize the run time available from those of us that are not constantly connected to the internet. Again, I want to thank to Robert for you help, its encouraging. ski1939 (team USMC) P.S. Email I sent Roman was returned as Delivery to the following recipients failed rm "AT" distributed.org.ru ---------------------------------------- [Edit 2 times, last edit by ski1939 at Feb 8, 2005 8:27:54 AM] |
||
|
|
RT
Master Cruncher USA - Texas - DFW Joined: Dec 22, 2004 Post Count: 2636 Status: Offline Project Badges:
|
I am at a loss. That looks like the correct email address. I suppose I would try it another time or two before I gave it up.
----------------------------------------Failing that, I think I would just wipe out the whole UDMon and all the slots. Then start over following these instructions . I may not have said it but, I used UDMON quite a bit a few years back (when I was on dialup) but was not an expert on it then and certainly am not now. I never did have any real problems with it though. Based on what you have said, you should remember that UDMon assumes that it starts and stops all slots(agents) and the only thing the user gets involved with is starting and stopping UDMon. Sorry I cannot be of any real help.  Good Luck! |
||
|
|
ski1939
Senior Cruncher Freeport, Illinois, USA Joined: Nov 17, 2004 Post Count: 209 Status: Offline Project Badges:
|
Hi Robert (RT) -
----------------------------------------Again Thanks for trying to help. I found another email address 'rm "AT" voices.ru' for Roman Mamedov but the email I sent came back again with some unintelligible response - - - - - - - "Failed to deliver to 'rm "AT" voices.ru' SMTP module(domain @xxx.xxx.xxx.xxx:voices.ru) reports: DNS Loop: A-record for voices.ru points back to us" - - - - - - - - As there is no email address associated with his ICQ 117667457 account, I tried to send him an online ICQ msg. I think after the current wu completes I'll take your suggestion and try installing the monitor again. Anyway, I want to thank you for your interest, helpful suggestions and encouragement. ski1939 (team USMC) ---------------------------------------- [Edit 1 times, last edit by ski1939 at Feb 8, 2005 8:30:55 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I found another email address 'rm@voices.ru' for Roman Mamedov but the email I sent came back again with some unintelligible response - - - - - - - "Failed to deliver to 'rm@voices.ru' SMTP module(domain @217.16.28.130:voices.ru) reports: DNS Loop: A-record for voices.ru points back to us" - - - - - - - - OK, what this response is saying is that the MTA (Mail Transfer Agent) is trying connect via SMTP (Simple Mail Transport Protocol) to an IP (Internet protocol) address of 217.16.28.130. It gets this address by using the DNS (Domain Name System). This translates domain names (such as voices.ru) to IP addresses. In addition, DNS has entries for what is called MX (Mail eXchanger) records. This tells the MTA which machine to contact to send email. Delving into the DNS, I can see the following for voices.ru: C:\WINDOWS\system32>nslookup Default Server: XXXX Address: w.x.y.z > set q=any > voices.ru Server: XXXX Address: wx.y.z Non-authoritative answer: voices.ru nameserver = ns1.kurepin.ru voices.ru nameserver = ns2.kurepin.ru voices.ru nameserver = ns3.kurepin.ru > server ns1.kurepin.ru Default Server: ns1.kurepin.ru Address: 217.16.28.129 > voices.ru Server: ns1.kurepin.ru Address: 217.16.28.129 voices.ru primary name server = ns1.kurepin.ru responsible mail addr = ruslan.kurepin.ru serial = 2000042451 refresh = 3600 (1 hour) retry = 600 (10 mins) expire = 86400 (1 day) default TTL = 3600 (1 hour) voices.ru nameserver = ns2.kurepin.ru voices.ru nameserver = ns3.kurepin.ru voices.ru nameserver = ns1.kurepin.ru voices.ru internet address = 217.16.28.130 ns1.kurepin.ru internet address = 217.16.28.129 ns2.kurepin.ru internet address = 81.195.245.98 ns3.kurepin.ru internet address = 62.141.70.98 > set q=MX > voices.ru Server: ns1.kurepin.ru Address: 217.16.28.129 voices.ru primary name server = ns1.kurepin.ru responsible mail addr = ruslan.kurepin.ru serial = 2000042451 refresh = 3600 (1 hour) retry = 600 (10 mins) expire = 86400 (1 day) default TTL = 3600 (1 hour) > ruslan.kurepin.ru Server: ns1.kurepin.ru Address: 217.16.28.129 ruslan.kurepin.ru canonical name = kurepin.ru kurepin.ru MX preference = 40, mail exchanger = mx2b.kurepin.ru kurepin.ru MX preference = 50, mail exchanger = mx3.kurepin.ru kurepin.ru MX preference = 5, mail exchanger = mx.exler.ru kurepin.ru MX preference = 10, mail exchanger = mx1.kurepin.ru kurepin.ru MX preference = 30, mail exchanger = mx2a.kurepin.ru kurepin.ru nameserver = ns.exler.ru kurepin.ru nameserver = ns1.kurepin.ru kurepin.ru nameserver = ns2.kurepin.ru kurepin.ru nameserver = ns3.kurepin.ru mx.exler.ru internet address = 217.16.18.193 mx1.kurepin.ru internet address = 217.16.28.129 mx2a.kurepin.ru internet address = 81.195.245.98 mx2b.kurepin.ru internet address = 195.68.165.130 mx3.kurepin.ru internet address = 62.141.70.98 ns.exler.ru internet address = 217.16.18.192 ns1.kurepin.ru internet address = 217.16.28.129 ns2.kurepin.ru internet address = 81.195.245.98 ns3.kurepin.ru internet address = 62.141.70.98 > What all this shows is that the email for voices.ru will actually be sent to ruslan.kurepin.ru ("responsible mail addr = ruslan.kurepin.ru") which is actually an alias for kurepin.ru ("canonical name = kurepin.ru"). kurepin.ru has 5 MX records associated with it. An MTA is supposed to take the MX record which has the lowest preference, which in this case is mx.exler.ru. In the example, you gave above, your MTA tried to contact mx2b.kurepin.ru which has preference 40, making it 4th in line. This would normally mean that there is a misconfiguration in the DNS, your MTA or the mail receiving MTA at the far end. Assuming that you can send email elsewhere, I would guess that it is either the DNS or the far end. Unfortunately, the serial number of the DNS recordsdoes not give many clues. The "standard" on the Internet is to use something like 2005070201, which would signify the first DNS records update on the 7th Feb, 2005. Hope this helps  Regarding the actual problem. I am running UD Monitor without any problems. However, if I just start UD Monitor, I get similar types of problems. You have to make sure that you launch UD Monitor through the included udmon-wcg.bat, or through a shortcut to "ud_mon.exe -wcg". |
||
|
|
RM
Cruncher Joined: Nov 16, 2004 Post Count: 5 Status: Offline |
Hello all, and sorry for the late reply.
----------------------------------------RM[stranger w/ 2 posts in 4 mos.]; I new that Sourceforge would be here [somewere since NOVEMBER!?!?!?] UD Monitor has nothing to do with Sourceforge, except for the fact that it is hosted there - just like many thousands of other open-source projects. For us 'NEWBIEZ, plz elaborate on what UDmonitor will do for the good of mankind [as we know it ]. Well, it will allow dial-up users to do more good(i.e. workunits) for WCG, which then, in turn, hopefully makes from these workunits something good for mankind(i.e. the cures).  What about GOOGle-labs F@H...? UD Monitor only supports the command-line version of F@h. It doesn't support the one built into the Google toolbar. ski1939 Regarding your problems with "another instance of the grid agent is already running". Can you stop the WCG Agent from UD Monitor? I mean, when the agent is running, UD Monitor displays "Running: Yes", and when you press the stop button, the Agent stops, and the label changes to "Running: No", right? OK, what this response is saying... Wow, a whole investigation :) I know that rm at distributed.org.ru or, preferrably, rm at distributed.ru should both work just fine. You have to make sure that you launch UD Monitor through the included udmon-wcg.bat, or through a shortcut to "ud_mon.exe -wcg". Yes, this is very important. [Edit 2 times, last edit by RM at Feb 7, 2005 6:39:28 PM] |
||
|
|
ski1939
Senior Cruncher Freeport, Illinois, USA Joined: Nov 17, 2004 Post Count: 209 Status: Offline Project Badges:
|
Hi Roman -
----------------------------------------Thanks for the reply and I apologize if I have unwittingly transgressed some rule unknown to me about putting an "AT" sign in a forum post. I was not aware that this should be avoided, but hopefully someone can enlighten me about the rule and the purpose of avoiding it in forum posts. I have never tried to stop the agent with the monitor and I am currently running the agent without the monitor (current wu is 86% at 44+ hours) so I am reluctant to try starting the monitor now. When I was getting the "another instance running" message I was setting up the cache slots and was being prompted to register the devices. I don't know what your suggestion "launch UD Monitor through the included udmon-wcg.bat" means or how to do it. I think I did set up a shortcut to "ud_mon.exe -wcg" and put it into startup properly, but now I'm not even sure of that. I guess the monitor is just too complex for me to setup and use correctly. However, I want to thank you again for your reply - ski1939 (team USMC) |
||
|
|
RM
Cruncher Joined: Nov 16, 2004 Post Count: 5 Status: Offline |
Thanks for the reply and I apologize if I have unwittingly transgressed some rule unknown to me about putting an "AT" sign in a forum post. I was not aware that this should be avoided, but hopefully someone can enlighten me about the rule and the purpose of avoiding it in forum posts. Spammers continiously harvest the web in search of new e-mail addresses - they do so not manually of course, but with automated tools, and these tools, as it is quite safe to suppose, catch only the addresses written in the 'plain' form, e.g. like "john@example.com". So if you do not take care to somehow obfuscate your address when writing it on the web (e.g. "john at example.com" or "johnATexampleDOTcom"), after some time you will start getting more and more offers to buy viagra, to enlarge certain body parts, or to help out the rich businessmen from Nigeria :) I don't know what your suggestion "launch UD Monitor through the included udmon-wcg.bat" means or how to do it. When you install UD Monitor, the destination folder contains the file named udmon-wcg.bat. If you participate in WCG, you should always run UD Monitor by double-clicking this file (or through a shortcut to "ud_mon.exe -wcg"), and never by simply launching ud_mon.exe. I think I did set up a shortcut to "ud_mon.exe -wcg" and put it into startup properly, but now I'm not even sure of that. If you didn't, i.e. created just a simple shortcut to "ud_mon.exe", that could explain all the problems with losing WUs which you were experiencing. |
||
|
|
|