Always "The services could not be started"

[shiliu]

FreeNAS version: 8.0.1 RC1, RC2, & Release
VM base: ESXi 4.1 Installable Update 1

There is no such issue with 8.0 Release. I newly installed these version of 8.0.1 for testing, but not able to start any services and got the error message "The services could not be started".

I tried to import the configuration of 8.0. It looks like works. But I cannot change the status of any services.

I also tried to reset to factory default, but still not work.


Do you have any ideas about this?

Thanks a lot

 

[shiliu]

Here is some info when I tried to start CIFS:

Sep 30 19:56:23 freenas freenas[1625]: Executing: /usr/sbin/service dbus forcestop
Sep 30 19:56:23 freenas freenas: dbus not running? (check /var/run/dbus/dbus.pid).
Sep 30 19:56:23 freenas freenas[1625]: Executing: /usr/sbin/service dbus restart
Sep 30 19:56:23 freenas freenas: Will not 'restart' dbus because dbus_enable is NO.
Sep 30 19:56:23 freenas freenas[1625]: Executing: /usr/sbin/service avahi-daemon forcestop
Sep 30 19:56:23 freenas freenas: Stopping avahi-daemon.
Sep 30 19:56:23 freenas freenas: Failed to kill daemon: No such file or directory
Sep 30 19:56:23 freenas freenas[1625]: Executing: /usr/sbin/service avahi-daemon restart
Sep 30 19:56:23 freenas freenas: Will not 'restart' avahi_daemon because avahi_daemon_enable is NO.
Sep 30 19:56:23 freenas freenas[1625]: Executing: /usr/bin/killall nmbd
Sep 30 19:56:23 freenas freenas: No matching processes were found
Sep 30 19:56:23 freenas freenas[1625]: Executing: /usr/bin/killall smbd
Sep 30 19:56:23 freenas freenas: No matching processes were found
Sep 30 19:56:23 freenas freenas[1625]: Executing: /usr/sbin/service samba quietstart
Sep 30 19:56:23 freenas freenas: Removing stale Samba tdb files: done
Sep 30 19:56:23 freenas freenas[1625]: Executing: /bin/pgrep -F /var/run/samba/smbd.pid smbd

 

[shiliu]

Can anyone help? I really want to use some new features in 8.0.1.

Thanks a lot.

 

[shiliu]

I CAN enable/disable all services yesterday without any change (at least I cannot remember any big change).

But I met this issue again today. I cannot enable/disable any service now.

Here is the action list what I did:
1. Create a new & only ZFS volume.
2. Create new users and a new group.
3. Update the admin user name and user password.

Here is some info when trying to disable AFP services:
Oct 5 12:31:31 freenas freenas[1746]: Executing: /usr/sbin/service ix-afpd quietstart
Oct 5 12:31:31 freenas freenas[1746]: Executing: /usr/sbin/service netatalk forcestop
Oct 5 12:31:31 freenas afpd[6827]: AFP Server shutting down on SIGTERM
Oct 5 12:31:31 freenas freenas[1746]: Executing: /usr/sbin/service dbus forcestop
Oct 5 12:31:31 freenas freenas: Stopping dbus.
Oct 5 12:31:31 freenas avahi-daemon[6809]: Disconnected from D-Bus, exiting.
Oct 5 12:31:31 freenas freenas[1746]: Executing: /usr/sbin/service dbus restart
Oct 5 12:31:31 freenas freenas: dbus not running? (check /var/run/dbus/dbus.pid).
Oct 5 12:31:31 freenas freenas: Starting dbus.
Oct 5 12:31:31 freenas freenas[1746]: Executing: /usr/sbin/service avahi-daemon forcestop
Oct 5 12:31:31 freenas freenas: Stopping avahi-daemon.
Oct 5 12:31:31 freenas freenas: Failed to kill daemon: No such file or directory
Oct 5 12:31:31 freenas freenas[1746]: Executing: /usr/sbin/service avahi-daemon restart
Oct 5 12:31:31 freenas freenas: Stopping avahi-daemon.
Oct 5 12:31:31 freenas freenas: Failed to kill daemon: No such file or directory
Oct 5 12:31:31 freenas freenas: Starting avahi-daemon.
Oct 5 12:31:31 freenas avahi-daemon[6959]: WARNING: No NSS support for mDNS detected, consider installing nss-mdns!
Oct 5 12:31:31 freenas freenas[1746]: Executing: /usr/sbin/service netatalk restart
Oct 5 12:31:31 freenas freenas: No matching processes were found
Oct 5 12:31:31 freenas freenas: No matching processes were found
Oct 5 12:31:31 freenas cnid_metad[6974]: Set syslog logging to level: LOG_NOTE
Oct 5 12:31:31 freenas [6976]: Set syslog logging to level: LOG_NOTE
Oct 5 12:31:31 freenas afpd[6977]: AFP/TCP started, advertising 192.168.16.110:548 (2.2.0)
Oct 5 12:31:31 freenas freenas[1746]: Executing: /bin/pgrep -F /var/run/afpd.pid afpd
Oct 5 12:31:31 freenas afpd[6977]: bind(fd, (struct sockaddr *)&address, address_length) failed: Address already in use



Here is the info when trying to disable CIFS:
Oct 5 12:32:35 freenas freenas[1746]: Executing: /usr/sbin/service dbus forcestop
Oct 5 12:32:35 freenas freenas: Stopping dbus.
Oct 5 12:32:35 freenas avahi-daemon[6959]: Disconnected from D-Bus, exiting.
Oct 5 12:32:35 freenas afpd[6977]: Failed to contact server: Daemon not running
Oct 5 12:32:35 freenas freenas[1746]: Executing: /usr/sbin/service dbus restart
Oct 5 12:32:35 freenas freenas: dbus not running? (check /var/run/dbus/dbus.pid).
Oct 5 12:32:35 freenas freenas: Starting dbus.
Oct 5 12:32:35 freenas freenas[1746]: Executing: /usr/sbin/service avahi-daemon forcestop
Oct 5 12:32:35 freenas freenas: Stopping avahi-daemon.
Oct 5 12:32:35 freenas freenas: Failed to kill daemon: No such file or directory
Oct 5 12:32:35 freenas freenas[1746]: Executing: /usr/sbin/service avahi-daemon restart
Oct 5 12:32:35 freenas freenas: Stopping avahi-daemon.
Oct 5 12:32:35 freenas freenas: Failed to kill daemon: No such file or directory
Oct 5 12:32:35 freenas freenas: Starting avahi-daemon.
Oct 5 12:32:35 freenas avahi-daemon[7084]: WARNING: No NSS support for mDNS detected, consider installing nss-mdns!
Oct 5 12:32:35 freenas freenas[1746]: Executing: /usr/bin/killall nmbd
Oct 5 12:32:35 freenas freenas[1746]: Executing: /usr/bin/killall smbd
Oct 5 12:32:35 freenas freenas[1746]: Executing: /usr/sbin/service samba quietstart
Oct 5 12:32:35 freenas freenas: Removing stale Samba tdb files: ....... done
Oct 5 12:32:35 freenas freenas: Starting nmbd.
Oct 5 12:32:35 freenas freenas: Starting smbd.
Oct 5 12:32:36 freenas freenas[1746]: Executing: /bin/pgrep -F /var/run/samba/smbd.pid smbd

 

[ProtoSD]

I tried to import the configuration of 8.0. It looks like works. But I cannot change the status of any services.

Can you try a clean installation without importing any settings and see if services work?

I think there are differences in the structure of the database between 8.0 and 8.01 and that could be causing problems for you. There was another problem reported about the database after restoring settings and doing a factory reset which might be related.

For now just try a clean installation and don't import any settings and see if it works.

 

[shiliu]

Thanks a lot for reply.

My 8.0.1 is a clean installation. In my current installation, I did not import any configuration. I just manually configure it. But still have this issue.

And I tried to do "Factory restore", still not work.

 

[ProtoSD]

Which version did you install i386 or AMD64?

What do you have for the physical hardware, cpu, memory, etc.?

Also - I sent you a PM, see your messages here on the forum.

 

[shiliu]

It is installed on ESXi 5.

Here is the detail info:
System Information

Hostname freenas.local
FreeNAS Build FreeNAS-8.0.1-RELEASE-amd64 (8081)
Platform Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz
Memory 4082MB
System Time Tue Oct 4 22:49:09 2011
Uptime 10:49PM up 33 mins, 0 users
Load Average 0.00, 0.01, 0.00
OS Version FreeBSD 8.2-RELEASE-p3

 

[flake]

same problem here. (and same logs)
i have two freenas 8.0.1
one just runs fine
the otherone is installed in a hyper-v virtual machine and is not able to start any services.

 

[ProtoSD]

I'm seeing several people with EXSi related issues, but not finding any solutions. I've looked but haven't been able to find any ideas for any of them yet.

@flake, you said one of yours runs just fine, what version of FreeNAS is it running, what are the system specs etc.?

 

[flake]

both are version 8.0.1 rc2.
the first one runs on following hardware:
mainboard: ASUS P8H67-V (Rev.3.0)
cpu: Intel Core i5-2500K PC1155 6MB Cache 3,3GHz retail
memory: KINGSTON 4GB DDR3 1333MHz CL9 Non ECC DIMM
power: LC-Power 650W LC6650 GreenPower3 14cm (80+)Vers.2.3
hd: 6 X Western Digital VelociRaptor 600GB, SATA 6Gb/s

the second freenas, that HAD the problem runs on hyper-v.
"had", because all of a sudden it works now.

i don't think what i changed has anything to do with fixing it, but i will list it.

what i did before was to replicate a snapshot from the first one to the second one:

zfs send volume0/dataset4@manual | ssh -i /data/ssh/replication 192.168.227.26 zfs receive -F backupvolume0/dataset4@manual

at that point in time (yesterday) i tried to start services, but was not able to.

today i wanted to destroy that snapshot on 2ndfreenas in the gui, but was not able to. the error message was "'backupvolume0/dataset0@manual': dataset does not exist" (it wanted to delete of dataset0 instead of dataset4)
so i manually destroyed that snapshot:

zfs destroy backupvolume0/dataset4@manual

and after that i tried to start services on 2ndfreenas with success.

 

[ProtoSD]

I don't use ESXi, but I'm trying to help find some clue to the problem. I did some searching yesterday, but no clear answers.

Thanks for posting back, maybe it'll help someone recognize where the problem is.

 

[exiled78]

I am also using ESXi 5. I cannot get the services to start with 8.02 or 8.01. I have tried building the VM as hardware version 7 as well as 8 (the default for ESXi 5). Any help would be greatly appreciated.

 

[William Grzybowski]

Currently freenas uses a non-deterministic way to determine whether a service has been started or not...
It issues the rc.d script, sleep for 0.5 and then checks for the .pid file and the existence of the process (this 0.5 time is to wait for the process to actually initialize and write the pid file)
Although it is non-deterministic and the proper solution would be using a file notifier at operating system level it seems to be working pretty ok expect for the users of ESXi. The reason for that is pretty unkown but i _think_ it is related to the use of a single vCPU (which is mandatory, due to a bug between esxi and freebsd).

So, right to the point, if anyone here is experience enough to give it a try:
mount -uw /
vi /usr/local/www/freenasUI/middleware/notifier.py

change the line 145 from
time.sleep(0.5)
to
time.sleep(2)

save it (:wq)

Restart django:
/usr/local/etc/rc.d/django restart

Then try to start services...

 

[exiled78]

Genius! That worked. 8.02 on ESXi working with the above mod :)

 

[exiled78]

I spoke to soon. It worked the first time, but now, despite several builds, I cannot get these services to start again. I don't know why it worked the first time I tried it. Frustrating!

 

[William Grzybowski]

Did you try increasing that value? to 5 perhaps?

 

[exiled78]

Yeah i tried up to 20.

 

[Obie]

I am having the same problem and also tried the fix posted by William with no results.

Does anyone else have any suggestions to deal with this issue?