Web Hosting Talk : Web Hosting Main Forums : Hosting Security and Technology : Cannot restart MySQL, HTTPD and server

[DelPierro]

I am not being able to restart the server, or MySQL or httpd via SSH.

PHP Code:

root@server [~]# /etc/init.d/httpd fullstatus

Looking up localhost
Making HTTP connection to localhost
Alert!: Unable to connect to remote host.

lynx: Can't access startfile http://localhost/server-status
root@server [~]# 


PHP Code:

root@server [~]# /etc/init.d/mysql reload
/etc/init.d/mysql: line 244: kill: (3375) - No such process
touch: cannot touch `/var/lib/mysql/server.lubnan.us.pid': Read-only file system 


Suggestions??

[activelobby4u]

The filesystem seems to be mounted as read only.

1. What does mount command on console show?
2. Are you able to touch/ create a file in your server?

[DelPierro]

Originally Posted by activelobby4u

The filesystem seems to be mounted as read only.

1. What does mount command on console show?

PHP Code:

root@server [~]# mount
/dev/mapper/VolGroup00-LogVol00 on / type ext3 (rw,usrquota)
none on /proc type proc (rw)
none on /sys type sysfs (rw)
none on /dev/pts type devpts (rw,gid=5,mode=620)
usbfs on /proc/bus/usb type usbfs (rw)
/dev/sda1 on /boot type ext3 (rw)
none on /dev/shm type tmpfs (rw)
none on /proc/sys/fs/binfmt_misc type binfmt_misc (rw)
sunrpc on /var/lib/nfs/rpc_pipefs type rpc_pipefs (rw)
/usr/tmpDSK on /tmp type ext3 (rw,noexec,nosuid,loop=/dev/loop0)
/tmp on /var/tmp type none (rw,noexec,nosuid,bind)
root@server [~]# 


2. Are you able to touch/ create a file in your server?

[/quote]

PHP Code:

root@server [~]# mkdir test
mkdir: cannot create directory `test': Read-only file system
root@server [~]# 


Thank you

[activelobby4u]

Originally Posted by DelPierro

PHP Code:

root@server [~]# mount
/dev/mapper/VolGroup00-LogVol00 on / type ext3 (rw,usrquota)
none on /proc type proc (rw)
none on /sys type sysfs (rw)
none on /dev/pts type devpts (rw,gid=5,mode=620)
usbfs on /proc/bus/usb type usbfs (rw)
/dev/sda1 on /boot type ext3 (rw)
none on /dev/shm type tmpfs (rw)
none on /proc/sys/fs/binfmt_misc type binfmt_misc (rw)
sunrpc on /var/lib/nfs/rpc_pipefs type rpc_pipefs (rw)
/usr/tmpDSK on /tmp type ext3 (rw,noexec,nosuid,loop=/dev/loop0)
/tmp on /var/tmp type none (rw,noexec,nosuid,bind)
root@server [~]# 


PHP Code:

root@server [~]# mkdir test
mkdir: cannot create directory `test': Read-only file system
root@server [~]# 


Thank you[/QUOTE]

Try:

mount -o remount,rw /

[linux-tech]

There's a REASON your filesystem is mounted as RO. Usually that's because

A> you have bad drives (possible)
B> you have reached critical mass on a partition (usually / or /var) and can't store any more

In order to resolve this, you need to find out which this is
To start with, assuming that the system is up, from ssh , type

df -m

If ANY of those show up as 100% used, then remount the FS as instructed (see previous posts) and clean it up

If NOT, then you need to reboot the server, and have an fsck run. Do NOT do this while the server is up and running, as this can and WILL cause filesystem issues.

mount -o remount,rw /

never just assume that it's ok to remount a file system. That will ONLY complicate issues. Address the issue properly by checking the partition and drives with fsck while the server is rebooting.

[DelPierro]

Thanks guys.

PHP Code:

root@server [~]# df -m
Filesystem           1M-blocks      Used Available Use% Mounted on
/dev/mapper/VolGroup00-LogVol00
                        232643     41724    179102  19% /
/dev/sda1                   99        16        79  17% /boot
none                       490         0       490   0% /dev/shm
/usr/tmpDSK                485        12       448   3% /tmp
/tmp                       485        12       448   3% /var/tmp
root@server [~]# 


None of my drives is 100% full. I can't reboot the server, neither via SSH nor WHM.

PHP Code:

n PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
16287 root      19   0  2332  572  504 R  100  0.1 300:58.93 shutdown 


[david510]

DelPierro,

Best option here is to go for an fask. Check the log files also to see any hardware failing errors.

[DelPierro]

Originally Posted by david510

DelPierro,

Best option here is to go for an fask. Check the log files also to see any hardware failing errors.

What's a fask? Would you please shed some lights on this.

I checked the error log files, and I don't see anything that could possibly mean a hardware failure.

[page-zone]

Best option is, as long as it is responding to commands - BACK UP STUFF.
Then worry about fsck'ing it. I assume you mean it won't do a shutdown from the comand line. Before doing a hard reset, I'd back stuff up.

[DelPierro]

I am not too familiar with Linux, could you please advise on a web site/tutorial that would walk me through the backup process?

[DelPierro]

Apparently, I can't backup as well.

PHP Code:

-bash: databasefile.sql: Read-only file system 


[david510]

Originally Posted by DelPierro

What's a fask? Would you please shed some lights on this.

Sorry for the mistake. I meant fsck (File System Check). Yes, if you can backup the data do it first and ask your host to perform an fsck.

[DelPierro]

Originally Posted by david510

Sorry for the mistake. I meant fsck (File System Check). Yes, if you can backup the data do it first and ask your host to perform an fsck.

Thanks.

I don't have the permission to backup, even though I am logged in as root. I am running fsck as I type.

[DelPierro]

I ran fsck, with no luck.

PHP Code:

root@server [~]# fsck
fsck 1.35 (28-Feb-2004)
e2fsck 1.35 (28-Feb-2004)
/dev/VolGroup00/LogVol00 contains a file system with errors, check forced.
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Pass 4: Checking reference counts
Pass 5: Checking group summary information
/dev/VolGroup00/LogVol00: 1031077/30261248 files (1.6% non-contiguous), 11630796/60506112 blocks
e2fsck 1.35 (28-Feb-2004)
/dev/sda1 is mounted.

WARNING!!!  Running e2fsck on a mounted filesystem may cause
SEVERE filesystem damage.

Do you really want to continue (y/n)? yes

/boot: clean, 44/26104 files, 19295/104388 blocks 


This is what I get, when I attempt to restart MySQL.

PHP Code:


root@server [~]# /etc/init.d/mysql restart
Shutting down MySQL/etc/init.d/mysql: line 220: kill: (3375) - No such process
...................................                        [FAILED]
rm: cannot remove `/var/lock/subsys/mysql': Read-only file system
Starting MySQL                                             [  OK  ]
root@server [~]# 


[page-zone]

If it's a cpanel server how old are the backups? If it's a server with paying customers you might want to hire someone instead of going through the learning curve with OPD (other peoples data). I think linux-tech in this thread is one such person.

rsyncing whole partitions to another drive would be what I'd try, if there are no recent backups. It may save you from a real crappy holiday season.