This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.
>From: William Boyd <address@hidden> >Organization: UCAR/MMM >Keywords: 200310071336.h97DaTk1024775 LDM IDD Bill, >Institution: NCAR/MMM >Package Version: 6.0.14 >Operating System: IRIX 6.5.10 >Hardware Information: SGI O200 >Inquiry: Described in Atachment 1. >LDM Problems RAP-MMM > >After RAP upgraded their LDM to ldm-6.0.14, MMM lost the feed from RAP. >MMM upgraded to ldm-6.0.14 and the problem continues. Celia and I >have not changed the configuration files. Debugging this problem >has brought up the issue that MMM is fed from RAP because iita.rap.ucar.edu >was housed in RAP. So I would like advice on debugging this problem >and allowing thelma to feed tea.mmm.ucar.edu would also help. The problem is not being caused by an upgrade of the LDM. Your tests below show that access to chisel by your machine is being blocked. This is most likely due to a firewall configuration either in RAP or in MMM. >>ping chisel.rap.ucar.edu >PING chisel.rap.ucar.edu (128.117.192.209): 56 data bytes >64 bytes from 128.117.192.209: icmp_seq=0 ttl=63 time=0.527 ms >64 bytes from 128.117.192.209: icmp_seq=1 ttl=63 time=1.519 ms >... > >>ldmping chisel.rap.ucar.edu >Oct 06 20:06:17 State Elapsed Port Remote_Host rpc_stat >Oct 06 20:06:17 ADDRESSED 0.010862 0 chisel.rap.ucar.edu Unable to re > ceive; errno = Connection reset by peer This shows that your LDM can not initiate contact with chisel. This is not an general problem on chisel since we can "see" it with an 'ldmping': -- on emo.unidata.ucar.edu -- /local/ldm% ldmping chisel.rap.ucar.edu Oct 07 14:55:41 State Elapsed Port Remote_Host rpc_stat Oct 07 14:55:41 RESPONDING 0.029618 388 chisel.rap.ucar.edu Oct 07 14:56:06 RESPONDING 0.000295 388 chisel.rap.ucar.edu Oct 07 14:56:31 RESPONDING 0.000389 388 chisel.rap.ucar.edu Oct 07 14:56:56 RESPONDING 0.002665 388 chisel.rap.ucar.edu >>notifyme -v -h chesel.rap.ucar.edu >ldmd.log >Oct 06 20:10:42 5Q:tea notifyme[65213]: Connected to upstream LDM-5 >Oct 06 20:10:42 3Q:tea notifyme[65213]: NOTIFYME(chisel.rap.ucar.edu): 7: Acce > ss denied by remote server If 'ldmping' fails, so will a 'notifyme'. >>cat ldmd.log >Oct 06 20:13:34 5Q:tea rpc.ldmd[65281]: Starting Up (version: 6.0.14; built: O > ct 2 2003 09:35:23) >Oct 06 20:13:34 5Q:tea chisel[65282]: Starting Up(6.0.14): chisel.rap.ucar.edu > .: TS_ZERO TS_ENDT {{ANY, ".*"}} >Oct 06 20:13:34 5Q:tea chisel[65282]: Desired product class: 20031006191334.21 > 8 TS_ENDT {{ANY, ".*"}} >Oct 06 20:13:34 3Q:tea chisel[65282]: ERROR: requester6.c:459; ldm_clnt.c:286: > nullproc_6 failure to chisel.rap.ucar.edu.; ldm_clnt.c:142: Unable to receiv > e; errno = Connection reset by peer This is another indication that your machine can not contact chisel. >Tests of different RAP server. > >>ping level.rap.ucar.edu >PING level.rap.ucar.edu (128.117.192.155): 56 data bytes >64 bytes from 128.117.192.155: icmp_seq=0 ttl=63 time=2.773 ms >64 bytes from 128.117.192.155: icmp_seq=1 ttl=63 time=1.836 ms >... > >>ldmping level.rap.ucar.edu >Oct 06 20:08:44 State Elapsed Port Remote_Host rpc_stat >Oct 06 20:08:44 ADDRESSED 0.012002 0 level.rap.ucar.edu Unable to rec > eive; errno = Connection reset by peer >>notifyme -v -h level.rap.ucar.edu >ldmd.log >Oct 06 19:54:30 5Q:tea notifyme[64860]: Connected to upstream LDM-5 >Oct 06 19:54:30 3Q:tea notifyme[64860]: NOTIFYME(level.rap.ucar.edu): 7: Acces > s denied by remote server Same problem here. The reason must be the same. Until we find the problem in RAP or MMM, please feed from from emo.unidata.ucar.edu. I just checked the allows on emo, and they are setup to feed any machine in the .mmm.ucar.edu domain. The only mmm machine that is currently allowed on thelma is box.mmm.ucar.edu. We will change the setup on thelma this afternoon (we have to time LDM stops and restarts on thelma to correspond to lulls in the CONDUIT datastream). Tom