- 39: ragwort, Site tower, Thu 07-Oct-2004 10:48:26 EDT, Ragwort Port powered up
Ragwort port powered back up at 10:50 local time.
- 31: ragwort, Site tower, Tue 31-Aug-2004 14:25:05 EDT, new ndaq code on ragwort
At 14:20 EDT a new version of the ndaq data acquisition code
was installed on the ragwort data system, and the ndaq restarted.
The new code has a fix which should cure the problem of
interrupted system calls when writing the local files.
Here's the error in the system log (which we're trying to cure):
IOException: /data/lrag_040825_000000.dat: write: Interrupted system call
- 26: ragwort, Site tower, Wed 25-Aug-2004 12:32:42 EDT, ragwort local storage stopped
The local storage process stopped at 4:20 this morning. Below is part
of the log messages on ragwort.
Aug 25 04:01:20 ragwort -- MARK --
Aug 25 04:03:12 ragwort ntpd[35]: synchronisation lost
Aug 25 04:20:10 ragwort Sampler[97]: RawSampleBuffer::distribute: IOException: /data/lrag_040825_000000.dat: write: Interrupted system call
Aug 25 04:21:22 ragwort ntpd[35]: offset 0.003465 sec freq 9.048 ppm error 0.010514 poll 4
Aug 25 04:41:20 ragwort -- MARK --
Aug 25 05:01:20 ragwort -- MARK --
Aug 25 05:21:20 ragwort -- MARK --
Aug 25 05:21:23 ragwort ntpd[35]: offset -0.002541 sec freq 13.066 ppm error 0.000894 poll 6
- 25: ragwort, Site tower, Tue 24-Aug-2004 10:26:10 EDT, Things to check on ragwort after power up
Here are a few things you should do after ragwort is powered back up. First
refer to entry 8 which explains how to bring ragwort back up after a power down.
Once ragwort is going do the following:
(1) "check_aster" to verify ingest, archive processes are running
(2) go to "/raw_data/all" to verify there is a new archive file
(3) do a "ndaqstatus ragwort" to check the data rates from ragwort
(4) log into root on ragwort: ssh root@ragwort (use the toto password)
(5) cd to /data and verify local storage is running
(6) do a "ps" to verify NDAQ process is running
- 18: ragwort, Site tower, Mon 16-Aug-2004 13:54:38 EDT, Reboot ragwort because of local storage
I rebooted ragwort because the local storage stopped on Sunday, Aug 15 13:03.
- 17: ragwort, Site tower, Thu 12-Aug-2004 16:33:20 EDT, Sensor on channel 218 causing buffer overflows
A sensor has been attached to channel 218 (/dev/ttyS2). It is probably
the wave height sensor, but something is either wrong in our
configuration or something's wrong with the sensor.
The ndaq program can't find the 0xfc end-of-record character
in the sensor, and so the read buffer overflows.
ndaq was complaining every time the buffer overflowed, which causes
a bunch of syslog processing and traffic over the net, since
all the messages are logged to the base system.
I think this caused the system clock on ragwort to be -5 seconds
wrong relative to the base system.
I installed a new version of the ndaq program on ragwort, which
counts the overflows, but only logs every 1000th overflow
in 5 minutes. ndaqstatus will also now report the overflow
counts.
Started the new ndaq code. Seems to be running OK. The data on
channel 218 is still screwey, but the system clock is OK.
- 8: ragwort, Site tower, Mon 02-Aug-2004 11:34:04 EDT, Ragwort shutdown /startup procedure when power tower is going down
At times power has to be shut off at the tower. When this takes place a notice
is sent out to the users letting them know in advance, usually 24 hours. This
will give us time to bring ragwort down in an orderly procedure. Once power
is restored they will let us know. Below are instructions to power up and down
ragwort.
Power Down:
(1) login into ragwort: ssh -l root 128.128.205.109
(2) type "halt". This will cause ragwort to disconnect from the login.
(3) go to WHOI web site: asitower.whoi.edu/ohats2/conf_ncar.html
(4) turn the details display on then select submit query
(5) login as ohats2
(6) the password is the aster password
(7) select the button to turnoff the 12 volt supply
(8) select configure port
(9) do an update status to verify the supply is off
Power up:
(1) go to WHOI web site: asitower.whoi.edu/ohats2/config_ncar.html
(2) turn on the details display then select submit query
(3) login as ohats2
(4) use the aster password
(5a) select the button to turn on port 18
(5b) select the button to turn on the 12 volt supply
(6) select configure port
(7) wait ~30 seconds for the power supply to turn on
(8) select update status only
(9) the display should show the 12 volt power is on.
(10) the current should be around 5.5 amps for 18 sonics and the data system
- 5: ragwort, Site shorelab, Tue 27-Jul-2004 13:23:37 EDT, ragwort ping example
64 bytes from ragwort (128.128.205.109): icmp_seq=1 ttl=64 time=0.691 ms
64 bytes from ragwort (128.128.205.109): icmp_seq=2 ttl=64 time=0.593 ms
64 bytes from ragwort (128.128.205.109): icmp_seq=3 ttl=64 time=0.637 ms
64 bytes from ragwort (128.128.205.109): icmp_seq=4 ttl=64 time=0.620 ms
64 bytes from ragwort (128.128.205.109): icmp_seq=5 ttl=64 time=0.615 ms
64 bytes from ragwort (128.128.205.109): icmp_seq=6 ttl=64 time=0.616 ms
64 bytes from ragwort (128.128.205.109): icmp_seq=7 ttl=64 time=0.612 ms
64 bytes from ragwort (128.128.205.109): icmp_seq=8 ttl=64 time=0.631 ms
64 bytes from ragwort (128.128.205.109): icmp_seq=9 ttl=64 time=0.612 ms
64 bytes from ragwort (128.128.205.109): icmp_seq=10 ttl=64 time=0.613 ms
64 bytes from ragwort (128.128.205.109): icmp_seq=11 ttl=64 time=0.617 ms
64 bytes from ragwort (128.128.205.109): icmp_seq=12 ttl=64 time=0.611 ms
64 bytes from ragwort (128.128.205.109): icmp_seq=13 ttl=64 time=0.622 ms
64 bytes from ragwort (128.128.205.109): icmp_seq=14 ttl=64 time=0.608 ms
64 bytes from ragwort (128.128.205.109): icmp_seq=15 ttl=64 time=0.610 ms
64 bytes from ragwort (128.128.205.109): icmp_seq=16 ttl=64 time=0.610 ms
64 bytes from ragwort (128.128.205.109): icmp_seq=17 ttl=64 time=0.608 ms
64 bytes from ragwort (128.128.205.109): icmp_seq=18 ttl=64 time=0.579 ms
64 bytes from ragwort (128.128.205.109): icmp_seq=19 ttl=64 time=0.607 ms
64 bytes from ragwort (128.128.205.109): icmp_seq=20 ttl=64 time=0.634 ms