Error - when restarting the vertica cluster node!!

[adrian.oprea]

Spread does not seem to be running on 10.xxxxxxx The database will not be started on this host.
Error: the vertica process for the database is running on the following hosts:
10.xxxxxxxx
10.xxxxxxxx
This may be because the process has not completed previous shutdown activities. Please
wait and retry again.

-when i start the spreadd
i get :
/etc/init.d/spreadd start
Starting spread daemon: spread death but the file pid exists

############ the db_log content ######################
Connected to spread at host 10.xxxxxxx(10.xxxxxxx) port 4803
[Fri 29 Jun 2012 16:57:57] SP_connect: DEBUG: Auth list is: NULL
[Fri 29 Jun 2012 17:03:57] SP_scat_receive: failed receiving header on session 5 (ret: 0 len: 0): Resource temporarily unavailable
Connected to spread at host 10.xxxxxxx(10.xxxxxxx) port 4803
[Fri 29 Jun 2012 17:09:33] SP_connect: DEBUG: Auth list is: NULL
Connected to spread at host 10.xxxxxxx(10.xxxxxxx) port 4803
[Fri 29 Jun 2012 17:11:47] SP_connect: DEBUG: Auth list is: NULL
#####################################################################

- i have restarted the machine , on the other 2 nodes is working !! only on this one after the node stop and start ! the spreadd service is not responding !!
Any ideas !!
THX

[adrian.oprea]

The error Is it possible because i have installed the cluster on Vertica CE version ? is it ?

[adrian.oprea]

Hy all !! after reading a bit on the forum i came to with the solution !!!

- it seams that the service could not find the files needed to start !! ?? why , i don't know !!
So after i have restarted the node !! checked the spread service : "ps -ef | grep spread " - got no result !!

- so i tried to stop the "non-running service" - and very strange the service "stoped" (i mean the service wasan't running but is stopped )Strange !!
-then i have issued the comand to start the spread service !!
:> /etc/init.d/spread start
--- and it started , next restarted the vertica service on the node and checked that database changes were made !!!
So problem solved !!

[JimKnicely]

Run the command ps -ef | grep -i spread on the node and make sure the spread software isn't running.

If it's not running then you should be able to safely delete the process file spread.pid and then start the spread software.

Typically the spread.pid file can be found in the /var/run/ directory.