time sync error

Hello guys,
i've got an urgent problem.
when agent login, phone rings (softphone sip) after 2 secs the browser shows time sync error.
yesterday we started working with vicidial in production.
i read serveral time sync posts, but nothing helps.
update_server_ip or restarted apache.
i'm using vicidbox 2.0.5 i only installed a sangoma PRI card and using 2 E1 lines.
phones / server everything has GMT 1, client workstation too.
strange, yesterday everything works fine....no changes.
only one thing was not normal, in the morning the e1 line was down.
i restarted the e1 line on our tk-system, after that, the vicidial server crashed (not pingbar anymore). i restarted the system and then i got the time sync error....
what can be the problem...its really important cause its now production
thanks for any help!!
regards
philip
EDIT:
i stopped asterisk, stopped wanrouter and modules.conf noload chan_zap.so and started asterisk...
the problem is still alive
i think its not the problem of the E1 lines
found something on /var/log/messages during boot
can this cause the problem? i've got this error since i install vicibox, its not new.
i found something interesting!!!
in the morning at 7.45 o clock the server crashed. i powered off the machine. in var/log/messages the time was 6.45 ... after boot (loading rc.local) the ntp starts and sync the time ... suddenly var/log/messages writes logs with the correct time (1 hour later)....
-> there must be the problem...but i don't find it
bios time was wrong (2 hours), fixed it, problem still alive
i've got an urgent problem.
when agent login, phone rings (softphone sip) after 2 secs the browser shows time sync error.
yesterday we started working with vicidial in production.
i read serveral time sync posts, but nothing helps.
update_server_ip or restarted apache.
i'm using vicidbox 2.0.5 i only installed a sangoma PRI card and using 2 E1 lines.
phones / server everything has GMT 1, client workstation too.
strange, yesterday everything works fine....no changes.
only one thing was not normal, in the morning the e1 line was down.
i restarted the e1 line on our tk-system, after that, the vicidial server crashed (not pingbar anymore). i restarted the system and then i got the time sync error....
what can be the problem...its really important cause its now production

thanks for any help!!
regards
philip
EDIT:
i stopped asterisk, stopped wanrouter and modules.conf noload chan_zap.so and started asterisk...
the problem is still alive
i think its not the problem of the E1 lines
found something on /var/log/messages during boot
- Code: Select all
Oct 2 08:45:30 ka-vici-dialer kernel: [ 171.435547] CPU: L2 cache: 1024K
Oct 2 08:45:30 ka-vici-dialer kernel: [ 171.435550] CPU: Physical Processor ID: 11
Oct 2 08:45:30 ka-vici-dialer kernel: [ 171.436012] CPU7: Intel(R) Xeon(TM) MP CPU 3.16GHz stepping 01
Oct 2 08:45:30 ka-vici-dialer kernel: [ 171.436031] Total of 8 processors activated (50716.15 BogoMIPS).
Oct 2 08:45:30 ka-vici-dialer kernel: [ 171.436386] ENABLING IO-APIC IRQs
Oct 2 08:45:30 ka-vici-dialer kernel: [ 171.436598] ..TIMER: vector=0x31 apic1=1 pin1=2 apic2=-1 pin2=-1
Oct 2 08:45:30 ka-vici-dialer kernel: [ 171.536748] ...trying to set up timer (IRQ0) through the 8259A ... failed.
Oct 2 08:45:30 ka-vici-dialer kernel: [ 171.536751] ...trying to set up timer as Virtual Wire IRQ... works.
Oct 2 08:45:30 ka-vici-dialer kernel: [ 171.755582] checking TSC synchronization [CPU#0 -> CPU#1]: passed.
Oct 2 08:45:30 ka-vici-dialer kernel: [ 171.775639] checking TSC synchronization [CPU#0 -> CPU#2]:
Oct 2 08:45:30 ka-vici-dialer kernel: [ 171.795646] Measured 637 cycles TSC warp between CPUs, turning off TSC clock.
Oct 2 08:45:30 ka-vici-dialer kernel: [ 171.795649] Marking TSC unstable due to: check_tsc_sync_source failed.
Oct 2 08:45:30 ka-vici-dialer kernel: [ 171.796017] Brought up 8 CPUs
can this cause the problem? i've got this error since i install vicibox, its not new.
i found something interesting!!!
in the morning at 7.45 o clock the server crashed. i powered off the machine. in var/log/messages the time was 6.45 ... after boot (loading rc.local) the ntp starts and sync the time ... suddenly var/log/messages writes logs with the correct time (1 hour later)....
-> there must be the problem...but i don't find it

bios time was wrong (2 hours), fixed it, problem still alive