"File Manager" keeps loading

  • wattaman
  • 09/25/08
  • Offline
Posted: Thu, 2011-04-21 06:15

Hi! Any idea why the "File Manager" doesn't load anymore? It was working fine and suddenly it stopped loading. It keeps spinning the java logo and that's it. I've tried in all browsers, also delete the browser's cache, cookies etc. My java is the last version, as it was this morning when the File Manager worked fine. I'm afraid is a server side issue.

P.S. I've tried from another computer but the applet still not loading


What happens if you restart

  • andreychek
  • 01/05/09
  • Offline
  • Thu, 2011-04-21 12:02

What happens if you restart Webmin (using "/etc/init.d/webmin restart") -- does that help?

-Eric


Unfortunately, nothing. Still

  • wattaman
  • 09/25/08
  • Offline
  • Sun, 2011-04-24 06:19

Unfortunately, nothing. Still not loading the application.


Maybe a security setting in

  • Locutus
  • 10/10/10
  • Offline
  • Sun, 2011-04-24 13:20

Maybe a security setting in your system got garbled and it's not "trusting" the Java applet anymore? Maybe clearing security config of the browser might help.


Nope, still nothing. I've

  • wattaman
  • 09/25/08
  • Offline
  • Sun, 2011-04-24 20:23

Nope, still nothing. I've re-checked and cleaned cookies and temp files for all the browsers I'm using, Chrome, IE, Firefox and Opera, all behave the same. I wonder if there's a way to deactivate/uninstall only the file manager in webmin and reinstall it again, maybe this "windows like solutions" will fix the problem.


You might turn on Debug

  • Locutus
  • 10/10/10
  • Offline
  • Mon, 2011-04-25 04:46

You might turn on Debug Logging in Webmin to see if it records any errors when you access the file manager. Also check the usual Linux logs for errors. It's rather rare that stuff fails under Linux without any trace or log entry. :)


I think I've found a clue,

  • wattaman
  • 09/25/08
  • Offline
  • Mon, 2011-04-25 06:17

I think I've found a clue, but not sure how to fix it. The last time I've used the file manager I've deleted, or so I thought, a folder on another hdd. I was thinking, that time, that the file manager just needs more time to delete it, but I think it actually timed out. Now, in the debug logs, there's the last line:

6796 [25/Apr/2011 13:03:59.111657] webminuser 115.87.208.143 mount CMD "cmd=e2label /dev/sdb1 2>&1"
I guess that there's a problem with the folder I was trying to delete or with the hdd, maybe. Also, under the hardware>partitions and local disks, when I click the second hdd I see this message, too: Errors logged 176 errors detected Of course, I have no idea how to fix it, if I'm right and is a hdd/partition issue. Is there a way to automatically check/fix disk errors, like on windows? I'm on centos 5.6, by the way. Also, unmounting and mounting the hdd doesn't help; I can unmount only using the force-unmount but when trying to mount again it just keeps loading.


Under Linux, the usual way to

  • Locutus
  • 10/10/10
  • Offline
  • Mon, 2011-04-25 12:12

Under Linux, the usual way to check a partition for errors is unmounting it (you have to do that, checking a mounted partition will most likely cause data damage) and then running fsck on it.

To check what the partition in question is used for, you can look for its entry in /etc/fstab.


I'm still trying to fix the

  • wattaman
  • 09/25/08
  • Offline
  • Mon, 2011-04-25 22:24

I'm still trying to fix the errors. I've made an error check for the second drive and it has 178 errors. The drive is unmounted and all the umount commands I've found and tried are telling me this. However, the fsck doesn't want to check/fix it saying the opposite: the drive is in use by something (?!) From within webmin (hdd management) I got the "Failed to save mount : Failed to lock file /etc/fstab after 5 minutes" when I try to program the drive for a check at the next reboot. I'll be using "shutdown -rF now" now, but I'm afraid it will only check the first hdd, the one where the system is installed. We shall see...


OK, the "shutdown -rF now"

  • wattaman
  • 09/25/08
  • Offline
  • Mon, 2011-04-25 22:57

OK, the "shutdown -rF now" (ssh) didn't work, nothing happens after the "The system is going to reboot..." message - I've waited half hour, I think the machine had should had enough time. Anyway, I'm just restarting now normally, from within webmin (system>bootup&shutdown), maybe some magic will happen and the system will remember the previous '-F' command and do the checking :) If not working, I'm afraid I'll adopt another radical solution and just reformat the drive :(


I'm sinking even deeper

  • wattaman
  • 09/25/08
  • Offline
  • Mon, 2011-04-25 23:20

I'm sinking even deeper LOL The shutdown command doesn't seem to work, either. Server won't restart. I'm doing a reboot from the hosting provider panel now.


New things happened after the

  • wattaman
  • 09/25/08
  • Offline
  • Tue, 2011-04-26 00:11

New things happened after the restart: - the hdd is not mounted, however from ssh it says it's mounted (when I run fsck); - same for the repair option under the webmin>hardware>partitions...etc; - trying to mount it again, from ssh or webmin, does nothing - ssh hangs and webmin "Failed to save mount : Failed to lock file /etc/fstab after 5 minutes"; - the file manager loads now;

Also, in the root folder '/' there a 'forcefsck' file that's empty - possible that the system tried the fsck after all but couldn't manage to fix it? Is there a way to perform the fsck and recover the drive? I know that if I tell the hosting company they'll just replace the hdd so I'll lose all the data there :(

Anyway, at least found out why the file manager was not loading.


Seems there is some serious

  • Locutus
  • 10/10/10
  • Offline
  • Tue, 2011-04-26 04:21

Seems there is some serious damage to the installation/filesystem.

I'd check the usual Linux log files in /var/log for any errors during boot/mount. This is important, really do it! I'm not making these suggestions just for fun.

Forget Webmin for now, do any repair attempts via the shell. An additional level of complexity and automatism like Webmin introduces is not recommended when things are damaged like they are.

What is your partitioning scheme? Post an out output of cat /etc/fstab and of mount after boot. (Enclose screen outputs in [code][/code] tags please for readability.)

To check if any files are still open on HDDs you want to unmount, use lsof | grep.

If checking the HDDs fails during boot, you can boot the machine from a rescue/installation CD. Hosters often offer such an option, to boot from a rescue system and mount your HDDs somewhere in the directory tree so that you can make tests without the system being active.

If all fails: you do have backups of your data, right? Ask your hoster to reinstall the system then and restore your backups.


There's a boot.log and a

  • wattaman
  • 09/25/08
  • Offline
  • Tue, 2011-04-26 05:58

There's a boot.log and a boot.log.1 but they're empty. The cat /etc/fstab will show:

/dev/sda1  /  ext3  grpquota,usrquota,rw,noatime  0  1
/dev/sda2         swap                    swap    defaults        0 0
tmpfs                   /dev/shm                tmpfs   defaults        0 0
devpts                  /dev/pts                devpts  gid=5,mode=620  0 0
sysfs                   /sys                    sysfs   defaults        0 0
proc                    /proc                   proc    defaults        0 0
/dev/sdb1  /HDD  ext3  defaults,noatime  0  0
mount /dev/sdb1 will do nothing, I must restart the putty again. The lsof | grep /HDD will show: bash: lsof: command not found

There's a rescue console I forgot about in my hosting panel, I'll try to boot from it and check the hdd right now. There's nothing very-very important on my second hdd, only photos and games, but they're a few GB and I'll lose another week or two to upload them again, if everything fails. Anyway, thanks for the help, I'll write how things will work after this.


boot.log should not really be

  • Locutus
  • 10/10/10
  • Offline
  • Tue, 2011-04-26 10:44

boot.log should not really be empty I think. For me, it lists the console output during boot.

There's more logfiles that might be relevant too: dmesg, debug, syslog, messages, kern.log.

Your installation does not have lsof? Okay, you might install the corresponding package, with yum or apt-get, depending on your distro.


Hi there! I've managed to

  • wattaman
  • 09/25/08
  • Offline
  • Sat, 2011-04-30 10:53

Hi there! I've managed to make it work. Well, actually I haven't done much, just restarted the server a thousand times until I've managed to unmount the drive. My hoster has some issues with the restart server and I couldn't use the recovery console. Anyway, the good news is that I've finally fsck-ed the drive and it repaired the problems. Now I'm trying to clean it from all the lost+found files it created. However, just wanted to thank you and to confirm that - or at least that's my noob conclusion - the file manager didn't load because of the second drive. Everything is fine now. Thanks again!


You're welcome, and great

  • Locutus
  • 10/10/10
  • Offline
  • Sat, 2011-04-30 16:40

You're welcome, and great that you got it fixed! :)