We ran into a situation a few days ago where all of our ptpd clients on our LAN shutdown simultaneously terminated (to within a second). They each showed the message "(ptpd notice) shutdown on terminate signal" at the end of the ptpd log, which seems to say that they caught a SIGTERM signal.
We're confused as to why they would shutdown simultaneously. Is there a means in which the master ptpd could signal them to terminate? Is it possible that a malformed packet could make them terminate?
We're running the ptpd-1.1.0.tar.gz package on 64bit centOS.
Any advice or suggestions for debugging this is most appreciated. Thanks.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The clients (slaves) were not terminating according to some message from the master ptpd, they were just not logging any csv data until they were terminated normally. The "shutdown on terminate signal" does not have a timestamp.
It appears that the reason that the master ptpd stopped sending out multicast packets was because the networking on the computer was restarted, e.g. /etc/init.d/networking restart
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I'd say this is not a bug then. I guess a feature request for time stamping the terminate message would be in order. Is that OK? If so please create a request for that.
Best,
George
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Done. In fact, I took advantage of the invitation to post three feature requests, which would solve the problems we have encountered so far in our installation.
Thanks for the support!
Cheers, -Russ
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
We ran into a situation a few days ago where all of our ptpd clients on our LAN shutdown simultaneously terminated (to within a second). They each showed the message "(ptpd notice) shutdown on terminate signal" at the end of the ptpd log, which seems to say that they caught a SIGTERM signal.
We're confused as to why they would shutdown simultaneously. Is there a means in which the master ptpd could signal them to terminate? Is it possible that a malformed packet could make them terminate?
We're running the ptpd-1.1.0.tar.gz package on 64bit centOS.
Any advice or suggestions for debugging this is most appreciated. Thanks.
Ok, here's an update…
The clients (slaves) were not terminating according to some message from the master ptpd, they were just not logging any csv data until they were terminated normally. The "shutdown on terminate signal" does not have a timestamp.
It appears that the reason that the master ptpd stopped sending out multicast packets was because the networking on the computer was restarted, e.g. /etc/init.d/networking restart
I'd say this is not a bug then. I guess a feature request for time stamping the terminate message would be in order. Is that OK? If so please create a request for that.
Best,
George
Done. In fact, I took advantage of the invitation to post three feature requests, which would solve the problems we have encountered so far in our installation.
Thanks for the support!
Cheers, -Russ