• Processing (Info)

    • MPC Status Page: Archive (2006 January-June)

    This page describes enhancements to or problems that have occurred with the MPC and scripts and the fixes that have been made.

    Recent problems are listed elsewhere. Index of other older problems..

    Older Enhancements and Resolved Problems

    • Overnight Problems
      2006 June 17: 12:00. Problems developed on two cluster members overnight, causing various processing problems. At this time it is not possible do a "clean" fix of the problems: a "dirty" fix has been implemented and some semblence of normality has returned. Further fixes will be done next week.

    • Overnight Problems
      2006 April 27: 11:00. A problem with erroneous elements being inserted into the secondary file used to prepare the MPCORB database occurred last night. The end result of this problem was that elements for numbered minor planets in the MPES were unavailable and the MPCORB database was blanked out. A problem with a full data disk also occurred. Both sets of problems are being fixed.

    • Multiple ACKs
      2006 April 22: 21:00. A problem that caused multiple ACKs to be sent has been cleared. We apologise for the multiple copies of ACKs that observers will have received for batches submitted between about 17:30 and 21:00 today.

    • Missing Orbit for (76390)
      2006 April 14: 12:11. Due to some over-zealous removal of erroneous orbits yesterday, the orbit for (76390) was inadvertently removed from the publically-visible orbit files when this morning's DOU MPEC was prepared. This had the effect in the Minor Planet Ephemeris Service of causing requests for numbered minor planet N to return the data for N+1, if N was larger than 76389. This problem did not affect the internal MPC orbit files. The orbit for (76390) has been reinserted into the publically-visible orbit files and normal behavior has been restored.

    • DOU MPEC
      2006 March 21: 11:08. Last night's DOU MPEC has been abandoned. Any files that get updated as a result of DOU issuance will not be updated. Normal schedule should be resumed tonight.

    • Intermittent Connectivity March 9
      2006 March 7: 14:53. We have been informed that Harvard will be performing network maintenance between 05:00 and 06:00 EST on Thursday, March 9. During this period there will be intermittent connectivity between the MPC computers and the outside world.

    • MPCORB Mirror
      2006 March 6: 14:59. The mirror site for the MPCORB datafile was last updated on Feb. 13. The non-updating since then has been traced to two problems: unexpected shutdown of the SSH service on the remote machine; and a problem with the user-key access caused by our conversion from using ftp to using sftp for copying jobs. With both these problems fixed, updating via the daily script should have recommenced. It hasn't. Running the script manually at the command line shows an error occurring. We have no idea what this error message means, so we are consulting with the CF.
      • 17:34. It seems that the cryptic error message indicated that the problem was a non-functioning sftp service at the remote node. The sysadmin of the mirror system has fixed the problem and we are currently doing a partial update of the mirror site. A full update will occur with the nightly DOU MPEC.

    • DOU MPEC
      2006 January 28: 08:58. Last night's DOU MPEC has been abandoned. Any files that get updated as a result of DOU issuance will not be updated. Normal schedule should be resumed tonight.

    • Connectivity to CF webserver restored
      2006 January 17: 17:12. Connectivity to the CF webserver has been restored via a new gateway machine. This connection should be much more reliable than the previous connection. A DOU MPEC has been issued in order to fully check the new connection. Examination of the log file is just starting, but assuming there are no problems, regular DOU MPEC will resume tonight.
      • 17:46. An examination of the log file of the afternoon DOU MPEC run shows a couple of minor problems, which we think we have fixed. Tonight's DOU MPEC should be issued as normal.

    • Connectivity down to CF webserver
      2006 January 15: 21:57. The gateway machine we use to transfer files to the CF webserver has hung. The Computation Facility has been notified: a reboot of the gateway machine is scheduled for tomorrow morning--if the hang is related to an earlier hardward problem the underlying problem may not get fixed until Wednesday. Tonight's DOU MPEC is being cancelled and jobs that update the webpages are being suspended until the hang is cleared or an alternate method of copying files is introduced.
      • January 16: 10:30. The affected machine has been rebooted. Assuming that it remains stable, DOU MPEC issuance will resume tonight.
      • It seems that the problem also affected the transfer of the observations for the latest mid-month MPS batch that are available via MPCAT-OBS service. Some observations have been reported missing from the numbered file: most of these observations are in the files associated with the last full batch, the remainder will reappear after the next mid-month MPS batch.
      • January 16: 21:30. The gateway machine has failed again. Issuance of tonight's DOU MPEC is doubtful...
      • January 16: 22:10. Although the CF gateway machine is back on-line, its continued connectivity cannot be guaranteed. It has been decided that tonight's DOU MPEC is canceled.

    • Download MPC/MPS/MPO batches
      2006 January 6: 13:40. Information has been posted on the Web Access to MPC/CBAT Computer Services regarding the arrangements that have been made to allow MPC subscribers to download PS/PDF versions of the various publications.

    • DOU MPEC
      2006 January 6: 08:30. Last night's DOU MPEC has been abandoned. Any files that get updated as a result of DOU issuance will not be updated. Normal schedule should be resumed tonight.

    • Erroneous Earth MOID value on MPEC 2006-A16
      2006 January 4: 20:08. MPEC 2006-A16 contains an erroneous value for the value of the Earth MOID. The value has been removed from the archived version of the circular (and hence from the HTML version on this website). A check of the library routine is being made to determine why this erroneous value was generated. Until a fix is in place, the Earth MOID for 2006 AQ displayed in, e.g., the MPES or in the lists of unusual minor planets, is to be ignored.
      • January 5: 12:50. An investigation shows that, as initially suspected, the problem affects only a very small percentage of objects (probably much less than 1 percent). We are still investigating a fix...
      • January 6: 19:30. It seems that the current method cannot be fixed for all situations. A totally new method of determining the MOID is being implemented. It should be on-line by tomorrow evening.
      • January 7: 18:00. We believe that the new MOID code just inserted into the MPES and other code seems to be stable and robust. Any lists containing MOIDs will be updated during their new routine update cycle.

    • Generalized Vaisala orbits for new designations in MPChecker
      2006 January 4: 00:10. A user reported that MPChecker was not listing recently-assigned designations. A check showed the objects concerned to be accessible via the MPES, but not the MPChecker. An investigation showed the relevant MPChecker file had not been updated since Dec. 23. A manual fix updated the file and a modification was made to the DOU MPEC to ensure the relevant updated procedure is run tonight (and that the run is properly logged!).
      • This problem also affected objects with non-Vaisala orbits. A number of orbits published previously on DOU MPECs will be republished on the next DOU MPEC in order to correct this problem.