CDLehner Posted March 22, 2011 Share Posted March 22, 2011 Never bumped into this before. 5-disc array was up and running, I installed a 6th disc outside the array, to run preclear against. I was running preclear, when in the dos window it said "lost connection to server". I tried to telnet back in, and I get "could not open connection to host, on port 23: connection failed". And now I can't hit the unRAID console either, from either this PC (which I rebooted) or another. NEVER seen this before. WTF? I don't want to hard shut-down, do I? CD Link to comment
CDLehner Posted March 22, 2011 Author Share Posted March 22, 2011 Does the console work? Nope; said that above. Help CD Link to comment
kizer Posted March 22, 2011 Share Posted March 22, 2011 I was going to say why not install SSH from unMENU since its Port 22, but I'm guessing you've lost connection. Can you ping your machine at all to see if it responds at all? ping xxx.xxx.xxx.xxx Link to comment
SSD Posted March 22, 2011 Share Posted March 22, 2011 If you get a nasty repeating series of errors, it can make the syslog grow until it consumes all available memory. All sorts of important processes will get killed to make more memory available until the machine crashes. You should open a telnet prompt and run the command: tail -f /var/log/syslog As errors are logged in the preclear you will see them in here. Hopefully when the computer crashes there will be hints in this telnet window to explain why. Link to comment
CDLehner Posted March 22, 2011 Author Share Posted March 22, 2011 I was going to say why not install SSH from unMENU since its Port 22, but I'm guessing you've lost connection. Can you ping your machine at all to see if it responds at all? ping xxx.xxx.xxx.xxx Can't remember if I'm on a dynamic or fixed addy; I can also ping the server name, yes? CD Link to comment
CDLehner Posted March 22, 2011 Author Share Posted March 22, 2011 If you get a nasty repeating series of errors, it can make the syslog grow until it consumes all available memory. All sorts of important processes will get killed to make more memory available until the machine crashes. You should open a telnet prompt and run the command: tail -f /var/log/syslog As errors are logged in the preclear you will see them in here. Hopefully when the computer crashes there will be hints in this telnet window to explain why. I suspect this may be what's going on. The disk I was running a preclear on, might entirely be "bad"; that's what I was checking to see. I had no idea that if it had enough errors, and they stacked-up in unRAID, it could crash the box? I'll run that command when I get home, but what exactly am I looking for? You make it sound inevitable that it will crash? What does that mean? There's no way to stop this, and get back online? CD Link to comment
kizer Posted March 22, 2011 Share Posted March 22, 2011 I don't see why you couldn't ping tower Never tried it honestly. LOL Link to comment
CDLehner Posted March 22, 2011 Author Share Posted March 22, 2011 I was going to say why not install SSH from unMENU since its Port 22, but I'm guessing you've lost connection. Can you ping your machine at all to see if it responds at all? ping xxx.xxx.xxx.xxx Nope; can't ping, and devices can't see the SMB shares. CD Link to comment
CDLehner Posted March 22, 2011 Author Share Posted March 22, 2011 If you get a nasty repeating series of errors, it can make the syslog grow until it consumes all available memory. All sorts of important processes will get killed to make more memory available until the machine crashes. You should open a telnet prompt and run the command: tail -f /var/log/syslog As errors are logged in the preclear you will see them in here. Hopefully when the computer crashes there will be hints in this telnet window to explain why. OK, what do you mean by a "telnet prompt"? I went to a command line, ran telnet, then put in tail^-f^/var/log/syslog and I get invalid command. Am I really just sitting here, waiting for the server to crash??? CD Link to comment
kizer Posted March 22, 2011 Share Posted March 22, 2011 If you get a nasty repeating series of errors, it can make the syslog grow until it consumes all available memory. All sorts of important processes will get killed to make more memory available until the machine crashes. You should open a telnet prompt and run the command: tail -f /var/log/syslog As errors are logged in the preclear you will see them in here. Hopefully when the computer crashes there will be hints in this telnet window to explain why. OK, what do you mean by a "telnet prompt"? I went to a command line, ran telnet, then put in tail^-f^/var/log/syslog and I get invalid command. Am I really just sitting here, waiting for the server to crash??? CD The only way you can really run what he said is if your actually at the command line prompt from tower. Link to comment
dgaschk Posted March 22, 2011 Share Posted March 22, 2011 And now I can't hit the unRAID console either, from either this PC (which I rebooted) or another. This console is a keyboard and screen physically attached to the server. I don't know what you mean when you say, "from either this PC (which I rebooted) or another." Link to comment
CDLehner Posted March 22, 2011 Author Share Posted March 22, 2011 And now I can't hit the unRAID console either, from either this PC (which I rebooted) or another. This console is a keyboard and screen physically attached to the server. I don't know what you mean when you say, "from either this PC (which I rebooted) or another." Oh; I thought he meant like http://tower (servername) So what, my next step is to attach a keyboard and monitor to the server? What should I be looking for? CD Link to comment
dgaschk Posted March 22, 2011 Share Posted March 22, 2011 A VGA (or DVI) monitor and a USB keyboard. EDIT: changed mouse to keyboard. No mouse needed. Link to comment
SSD Posted March 23, 2011 Share Posted March 23, 2011 If you get a nasty repeating series of errors, it can make the syslog grow until it consumes all available memory. All sorts of important processes will get killed to make more memory available until the machine crashes. You should open a telnet prompt and run the command: tail -f /var/log/syslog As errors are logged in the preclear you will see them in here. Hopefully when the computer crashes there will be hints in this telnet window to explain why. OK, what do you mean by a "telnet prompt"? I went to a command line, ran telnet, then put in tail^-f^/var/log/syslog and I get invalid command. Am I really just sitting here, waiting for the server to crash??? CD Telnet is a protocol used to run a command line session, usually to a remote machine. Most people use the telnet client "putty" here (you can find a link to it on the Best of the FOrums - see my sig.) With telnet running in a window on your workstation, you can run the tail command I mentinoed (no ^ or other funky characters though) inside that putty window. When new rows are written to the syslog file, they will automatically show up in your putty window. You said that every time you run preclear on this disk your computer crashes. My guess is the crash is due to excessive logging. If you run the preclear while the tail command is running on your desktop, you will see the log entries as they occur. And if you are not watching, you'll be able to see a screenful of the most recent log entries. If the computer crashes, the putty windows will still show you that last screenful. Hope this makes more sense! Link to comment
dgaschk Posted March 23, 2011 Share Posted March 23, 2011 If you get a nasty repeating series of errors, it can make the syslog grow until it consumes all available memory. All sorts of important processes will get killed to make more memory available until the machine crashes. You should open a telnet prompt and run the command: tail -f /var/log/syslog As errors are logged in the preclear you will see them in here. Hopefully when the computer crashes there will be hints in this telnet window to explain why. OK, what do you mean by a "telnet prompt"? I went to a command line, ran telnet, then put in tail^-f^/var/log/syslog and I get invalid command. Am I really just sitting here, waiting for the server to crash??? CD Telnet is a protocol used to run a command line window, usually to a remove machine. Most people use the telnet client "putty" here (you can find a link to it on the Best of the FOrums - see my sig.) With telnet running in a window on your workstation, you can run the tail command I mentinoed (no ^ or other funky characters though) inside that putty window. When new rows are written to the syslog file, they will automatically show up in your putty window. You said that every time you run preclear on this disk your computer crashes. My guess is the crash is due to excessive logging. If you run the preclear while the tail command is running on your desktop, you will see the log entries as they occur. And if you are not watching, you'll be able to see a screenful of the most recent log entries. If the computer crashes, the putty windows will still show you that last screenful. Hope this makes more sense! Telnet uses port 23. Since you can't telnet you'll need a console. Link to comment
CDLehner Posted March 23, 2011 Author Share Posted March 23, 2011 A VGA (or DVI) monitor and a USB keyboard. EDIT: changed mouse to keyboard. No mouse needed. OK...but what should I be looking for once I have them hooked-up? EDIT- Ok, server is at login. What now? CD Link to comment
dgaschk Posted March 23, 2011 Share Posted March 23, 2011 You'll be able to copy the syslog to the flash drive. Enter: cp /var/log/syslog /boot/syslog.txt To restart the web interface: killall emhttp nohup /usr/local/sbin/emhttp & Then you can shutdown and move the USB drive to a PC in order to post the syslog. Link to comment
SSD Posted March 23, 2011 Share Posted March 23, 2011 You'll be able to copy the syslog to the flash drive. Enter: cp /var/log/syslog /boot/syslog.txt To restart the web interface: killall emhttp nohup /usr/local/sbin/emhttp & Then you can shutdown and move the USB drive to a PC in order to post the syslog. My technique is from a fresh boot when you know a preclear is going to crash the machine. If you can salvage the existing syslog, as described here, that is better. Link to comment
CDLehner Posted March 23, 2011 Author Share Posted March 23, 2011 You'll be able to copy the syslog to the flash drive. Enter: cp /var/log/syslog /boot/syslog.txt To restart the web interface: killall emhttp nohup /usr/local/sbin/emhttp & Then you can shutdown and move the USB drive to a PC in order to post the syslog. Copy seemed to work, but the killall gives me no process killed? Just to clarify...is this 2 commands; killall emhttp, then nohup? or all one command? CD Link to comment
dgaschk Posted March 23, 2011 Share Posted March 23, 2011 killall did nothing because it's already dead. Just do it to make sure. Enter this as a single line: nohup /usr/local/sbin/emhttp & Link to comment
CDLehner Posted March 23, 2011 Author Share Posted March 23, 2011 killall did nothing because it's already dead. Just do it to make sure. Enter this as a single line: nohup /usr/local/sbin/emhttp & OK, and then what? I can power down, pull the usb, and try to get at the syslog? OK...I ran nohup /usr/local/sbin/emhttp & and I get "ignoring input and appending output to 'nohup.out'" CD Link to comment
SSD Posted March 23, 2011 Share Posted March 23, 2011 killall did nothing because it's already dead. Just do it to make sure. Enter this as a single line: nohup /usr/local/sbin/emhttp & OK, and then what? I can power down, pull the usb, and try to get at the syslog? OK...I ran nohup /usr/local/sbin/emhttp & and I get "ignoring input and appending output to 'nohup.out'" CD If emhttp is running, you should be able to access the Web GUI. And from there stop the array and reboot. Then you should be able to access the flash share, and locate the saved syslog (which you should zip and post here). Link to comment
CDLehner Posted March 23, 2011 Author Share Posted March 23, 2011 killall did nothing because it's already dead. Just do it to make sure. Enter this as a single line: nohup /usr/local/sbin/emhttp & OK, and then what? I can power down, pull the usb, and try to get at the syslog? OK...I ran nohup /usr/local/sbin/emhttp & and I get "ignoring input and appending output to 'nohup.out'" CD If emhttp is running, you should be able to access the Web GUI. And from there stop the array and reboot. Then you should be able to access the flash share, and locate the saved syslog (which you should zip and post here). I don't know that emhttp is running; I can only tell you what happens when I put in that nohup command. I cannot get to the web GUI. CD Link to comment
dgaschk Posted March 23, 2011 Share Posted March 23, 2011 If you still can access the web interface. You'll have to shutdown another way. Do you have the powerdown addon installed? If not see this post: http://lime-technology.com/forum/index.php?topic=1905.msg13461#msg13461 Link to comment
Recommended Posts
Archived
This topic is now archived and is closed to further replies.