Re: [ibr-dtn] DTN bundle rejected issue
The API error is just a debugging message and is to be expected when a client (dtnping) disconnects form the dameon when it terminates. For your observation, you probably have some problem with discovery. It is hard to guess what exactly is wrong in your setup but the behavior can be explained like this:
A sees B, but B does not see A (i.e. because sending or receiving of A's discovery beacons is blocked). In this case B can not initiate a connection to A (dtnping not possible), but A can dtnping B. Once a TCPCL connection is established by A however, B becomes aware of A's existence as a neighbor, and thus it will also be able to contact A.
MfG
Sebastian
Am 05.03.2013 um 13:41 schrieb Francisco Cuesta ndarkness@gmail.com:
Sebastian,
thanks for your help, it was indeed an issue with time zones, one was working with GMT and the other with UTC, I have set both to UTC and now it seems to work. Though, sometime one of the nodes, the C, where I got the error I showed before (the API one) the dtnping doesn't always work properly I have to make a dtn ping from the other side and then the dntping works on the node C, any idea of this behaviour?
Thanks again,
Regards
2013/3/5 Sebastian Schildt schildt@ibr.cs.tu-bs.de
The default timeout for dtnping is 60 seconds. If your clocks are synchronized within one minute and you still have problems, I guess you have a problem with timezones. Normally, both hardware clocks should be set to UTC and the systems can use any timezone (even different ones). However, especially with some embedded hardware it is not always quite clear whether the hw clock is set to UTC or local time and how timezones are implemented.
You can try by increasing the lifetime in dtnping, i.e.
dtnping --lifetime 7260 dtn://destination/echo
would make the bundle life for 2 hours and 60 sec, thus by playing with lifetime you can find wether there is a timezone problem, and adjust the clocks on your nodes accordingly.
MfG
Sebastian
Am 05.03.2013 um 09:17 schrieb Francisco Cuesta ndarkness@gmail.com:
Good morning,
Looking into ibrdtn mail list, I have found a thread in which Johannes explained some way around to avoid the issues with the clock synchronization. So what I did was to run the daemon with the option --badclock and now the dtnping works fine as well as sending stuff through dtn nodes(just in one way from node 2 to node C); however, I would like to solve the issue with the clocks, could you tell me what is the precision that I would achieve with the clocks? Since, I have already set both dates on the systems with maybe a difference of few seconds, but if I disable the --badclock option then I still have that error.
On the other hand, I have noticed that from one node I cannot send a ping or send data, when I try to send data I was prompted the below messages. Any tip of what's going on?
BinaryStreamClient::run(): InvalidDataException (Bundle version differ from ours.) Tue Mar 5 08:15:56 2013 DEBUG.10: unexpected API error! Conditional has been unblocked. Tue Mar 5 08:15:56 2013 DEBUG.5: api connection down
Thanks in advance,
Francisco
2013/3/4 Francisco Cuesta ndarkness@gmail.com Hello,
I'm trying to do a dtnping between two dtn nodes I have, I have noticed that they can see each other, however, when I try to make a ping from node 2 to node C, I get this warning :
Mon Mar 4 13:49:45 2013 WARNING: bundle rejected: bundle has expired ([415720186.377] dtn://node2.dtn/zsKZSPktexxyNvJ -> dtn://nodeC.dtn/echo) Mon Mar 4 13:49:45 2013 WARNING: bundle has been rejected: bundle is expired
I see the packages is discarded, but I cannot understand the reason, is it due to a bad synchronization among the clocks of the devices? How may I resolve it?
Thanks in advance,
regards!
-- !! This message is brought to you via the `ibr-dtn' mailing list. !! Please do not reply to this message to unsubscribe. To unsubscribe or adjust !! your settings, send a mail message to ibr-dtn-request@ibr.cs.tu-bs.de !! or look at https://www.ibr.cs.tu-bs.de/mailman/listinfo/ibr-dtn.
-- !! This message is brought to you via the `ibr-dtn' mailing list. !! Please do not reply to this message to unsubscribe. To unsubscribe or adjust !! your settings, send a mail message to ibr-dtn-request@ibr.cs.tu-bs.de !! or look at https://www.ibr.cs.tu-bs.de/mailman/listinfo/ibr-dtn.
participants (1)
-
Sebastian Schildt