• Processing (Info)

    • MPC Status Page: Archive (2011 January-December)

    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

    • December 31 DOU MPEC
      2011 Dec. 31: 15:54. The December 31 DOU MPEC was abandoned due to a system failure early on in the production procedure.

    • Network Outage Planned (Monday Dec. 19)
      2011 Dec. 15: 20:50. We have just been informed that there will be a four-hour network outage at CDP (where the MPC is now based) from 23:00 UTC (18:00 EST) on Monday, December 19. This outage is to allow the building landlord to re-route network cables into the building. The outage will affect access to the MPC website and web scripts. The AUTOACK process will also be disabled during this outage.

    • New Object Ephemeris Generator
      2011 Nov. 29: 23:05. A corrected executable has been put in place for the New Object Ephemeris Generator script. The long-standing problem of ignoring a user's motion options has been fixed.

    • Problem with MPC Database
      2011 Nov. 27: 15:30. There is a problem with the MPC DB search feature. This is related to additional information that is being incorporated into the DB. The relevant person has been alerted.
      • 16:35. Normal functioning of the MPC Database has been restored.

    • Problem with Lightcurve Database
      2011 Nov. 26: 03:00. Following a report from a user, it is apparent that there is a problem with the software controling the Lightcurve Database. The relevant person has been alerted.
      • 03:55. Normal functioning of the Lightcurve Database has been restored.

    • Forthcoming E-Mail Outage (Friday Nov. 25)
      2011 Nov. 24: 03:30. We have just been informed that the Computation Facility will be installing OS patches between 11:05 UTC and 11:35 UTC on Friday, Nov. 25. During this time period e-mail messages will not be delivered (either inbound or outbound). Incoming e-mail should be redelivered automatically once the work is completed.

    • Minor Planet Ephemeris Service not functional
      2011 Nov. 14: 15:45. A number of users have reported that the MPES has been non-functional since 06:30. We are investigating.
      • 15:57. An examination of the script-server error logs shows that the first instance of this problem occurred at 06:17:53 UTC.
      • 16:23. Some garbage was inadvertantly added to one of the monthly work files. This garbage has been removed from both the master monthly work files and from the versions used in the MPES. The MPES should now return ephemerides as normal.

    • Incorrect Dates of Discovery and Sites for Numbered Minor Planets
      2011 Nov. 11: 20:26. Following a report from a user about apparent incorrect discovery circumstances for newly-numbered minor planets, we are checking the details for 107 suspect objects.
      • Nov. 11: 22:35. A number of published discovery details will require modification. We will make the changes directly in the master discovery files and document those changes in the next batch of Minor Planet Circulars. We are also investigating why the problem occurred.
      • Nov. 12: 17:35. The reason for the problem has now been determined and the documentation for the preparation procedure has been modified to better handle the issue. We can now investigate what changes to the discovery details need to be made.
      • Nov. 24: 03:25. The discovery details for 77 numbered minor planets have been corrected in the on-line list. Corrected details will appear in the December batch of Minor Planet Circulars.

    • November 9 DOU MPEC
      2011 Nov. 9: 11:00. The November 9 DOU MPEC (2011-V33) is being mailed out. It is large (~5.6 MB) and this is causing many mail servers to refuse delivery. Given the size of the circular, it may take some time for the mailing process to complete. MPEC 2011-V33 is the largest such circular ever issued and is almost twice the size of the previous "record holder", MPEC 2001-W71.

    • November 8 DOU MPEC
      2011 Nov. 8: 13:00. The November 8 DOU MPEC was abandoned following a problem with production caused by the huge number of orbits.

    • Missing Objects in MPChecker
      2011 Oct. 24: 18:00. A user reported that entering an observation of a newly-designated minor planet into the MPChecker did not produce a match to that object. The problem was traced to the build procedure of the data files for the webserver. This problem has existed since Oct. 11. The problem has been fixed and updated datafiles have been copied over to the webserver.

    • Väisälä Orbits in MPES
      2011 Oct. 14: 15:30. We have received a couple of reports of problems with Väisälä orbits in the MPES. The first and last dates are shown as -4712 Jan. 1 and the absolute magnitude is zero. These quantities appear to have been trashed on the orbit records for some Väisälä objects on the webserver. Not all objects are affected. We are investigating.
      • 15:50. We have identified the problem, an issue with building the secondary datafiles for the webserver. The fix will be visible externally after the next DOU MPEC. In the meantime, the ephemeris positions of the affected objects are correct.

    • Publication of 2011 Oct. 12 MPCs
      2011 Oct. 13: 15:30. A couple of problems have surfaced with the on-line publication of the 2011 Oct. 12 MPCs. The protection on the complete file of observations of numbered minor planets was incorrect, so that outside users could not access it. In addition, the file itself was truncated (being only about half the correct length). The first problem may have been a consequence of the second. Both problems have been fixed. Unfortunately, there are no log files available to check, since the file was prepared on a Linux system and transferred to another Linux system. So we have no idea what went wrong and we cannot be sure that this problem will not reoccur in the future.

    • Forthcoming E-Mail Outage (Wednesday Oct. 19)
      2011 Oct. 12: 16:03. We have just been informed that the Computation Facility will be performing disk upgrades between 21:30 UTC and 23:00 UTC. It is unclear if this maintenance will affect SMTP e-mail delivery. If it does, messages will not be delivered during the above-mentioned period, but should be redelivered automatically once the work is completed.

    • Repeated Observations in DataBase
      2011 Sept. 15: 19:05. A user reported finding duplicate observations when extracting observations via the DataBase interface. Strictly speaking, these are not duplicate observations as the records differ in the reference. In each case, one of the observations appeared on an MPEC, while the other appeared on a MPS. Clearly, the MPEC-published observations are not being replaced by the MPS-published observations. We are examining this problem. The duplicate observations are not present in our master files (copies of which are available via the MPCAT-OBS service). The duplicate observations in the DB will be removed shortly. Another problem that we will fix is related to the insertion of some observations that do not seem to have been inserted correctly.

    • Forthcoming Network Outage (Sunday Sept. 18)
      2011 Sept. 15: 18:25. We have just been informed that the Computation Facility will be performing network maintenance between 11:00 and 12:00 UTC on Sept. 18. Connections to our web services will be unavilable for up to 30 minutes in this window.
      • Sept. 18: 13:50. Automated processes and execution queues were shutdown beginning at 10:00 to try and prevent problems with remote resources being inaccessible. All execution queues have been restarted and the automated processes are restarting.

    • Planned Outage in AUTOACK
      2011 Sept. 15: 14:30. There will be a brief outage in the AUTOACK service between 16:00 and 17:00 UTC. This is to allow a NIC to be replaced. AUTOACK is the automated process that accepts incoming e-mail and acknowledges observation batches. Assuming no problems with removing the old NIC, downtime should be about 10 minutes. No special action is required by submitters of observations.
      • 17:05. This work has been postponed due to the need for further examination of network errors.

    • Problem with MPES and Comet Fragments
      2011 Aug. 20: 18:30. A user reported that requesting "213P" from the MPES resulted in the elements/ephemeris for "213P-B" being returned. This is now fixed.

    • "Customize This Page" Problem
      2011 Aug. 15: 22:20. A user reported errors when attempting to use the Customize This Page script. A simple recopy of the executable fixed the problem.

    • July 26 DOU MPEC
      2011 July 26: 14:00. The July 26 DOU MPEC was abandoned following the unplanned reboot of a cluster machine overnight.

    • Problem With Designation Linking in MPC Database
      2011 July 22: 22:22. The identification/double-designation table in the MPC Database is corrupted and will have to be rebuilt. This rebuild will occur in the near future and will mean that DB queries using non-principal designations may not return the expected results until such time as the table is rebuilt completely. Some small insertions into the new table will be used to check that all is well before the main bulk of the identiications are inserted. We apologize for the inconvenience.
      • July 23: 16.00. The test insertions into the new table showed the same problem. A close inspection of the relevant script revealed the reason for the mis-linkings. A fix is in the works, then the table will be rebuilt anew.
      • July 23: 18.20. The identification/double-designation table has been rebuilt. Tests show the problem cases identified earlier are now handled correctly. Please report any anomalies.

    • Problem With Script Server
      2011 July 18: 15:00. A number of users have reported problems accessing the webscripts. We are investigating.
      • 16:00. The problem was traced to a malformed data file. It was replaced and a number of zombie jobs were deleted.

    • Problem With Certain Objects in MPES
      2011 July 16: 15:30. A user has reported problems accessing ephemerides for unnumbered minor planets in the MPES. The problem is due to the non-instantaneous nature of the updating procedure associated with the filing of material from a MPC batch. The last bits of the filing are being performed now.
      • 18:20. Generation and copying of data files is complete.

    • Protection on Complete File of Numbered Minor Planet Observations
      2011 June 20: 15:50. A user reported an Error 403: Forbidden error when trying to access the complete file of numbered minor planet observations. It seems that the copy procedure that placed the files in that directory set the protection for that one file (alone!) to no access for anyone other than the owner. All the other files transferred at the same time had correct protections. The protection has been reset to read access for all.

    • Problem With Certain Objects in MPES
      2011 June 16: 18:15. A number of users reported problems accessing ephemerides for objects with Väisälä elements. A request for 2011 LN3, produced output for 2011 KD41. The problem, now fixed, was traced to a failure to unpack a zip file containing updated index files for the orbital elements. This was a transient problem. The sftp system does not report a useful error message to the OS. A workaround for this glaring limitation has been implemented.

    • Shutdown of www.minorplanetcenter.org
      2011 June 7: 16:50. The www.minorplanetcenter.org site has been shutdown. Traffic to that site will now be redirected to the .net site. It may take some time for the redirect to make its way around the name servers.

    • Network Reconfiguration
      2011 June 4: 13:00. There may be temporary outages in the web services over the next hour or so while we do some reconfiguration of the network connections for the public-facing web machines.

    • MPES Problem
      2011 May 11: 15:00. We have received a number of reports that the MPES is returning incorrect orbits for objects designated in 2011. We are investigating.
      • 15:55. The problem was traced to a .zip file copied to the script server that was not unpacked. Normal service should have been resumed.

    • Ubasti Retired
      2011 May 2: 18:00. We have been informed that access to the http port on the old webserver (Ubasti) has been disabled by the Computation Facility. Please use the corresponding pages on the .net site. With one exception, all services have been ported to the new server. The one exception is the MPCOBS service: users should extract the observations they require from the MPC Database interface.

    • New Object Ephemeris Generator
      2011 Apr. 30: 22:15. We have received a couple of reports of problems with the new New Object Ephemeris Generator script. These include: misreading the date if no date is entered and hourly output of ephemerides is selected (fixed); incorrect motions (the motions are always displayed in "/hour, regardless of the selection); and incorrect local circumstances. Fixes are being worked on.

    • Incorrect Dates of Discovery and Sites for Numbered Minor Planets
      2011 Apr. 21: 21:25. Following a report from a user about apparent incorrect dates of discovery for four recently-numbered minor planets, we performed a check to uncover five additional cases. The objects affected are (264210), (266779), (266959), (266966), (266976), (275353), (275405), (275426) and (275478). The discovery site (and the discoverer in one case) for (264235), (266971), (273868) and (275212) require correction. All these corrections will be documented fully in the next MPC batch, but the changes have been made in our internal files and in the files that are accessible through the website.

    • E-mailed Problems with MPECs
      2011 Apr. 21: 20:00. A number of users have reported non-receipt of recent MPECs. We don't see anything obviously wrong at this end, so the problem may be something transient with the CfA network. We will monitor the situation.

    • Problem with Minor Planet Designation Converter
      2011 Apr. 16: 14:00. A user has reported that the Minor Planet Designation Converter is returning results for a different object than that requested. The problem is due to a mismatch between the index file and data file that contains the information. This problem will be examined after the filing for the April MPCs is completed, so that fresh copies of the relevant files are available for examination.

    • SMTP E-mail Interruption (March 31)
      2011 Mar. 30: 13:45. We have been informed that the CF's SMTP server will be off-line for approximately 15 minutes beginning at 10:05 UTC on Thursday, March 31, to allow OS patches to be installed. Attempts to send e-mail to us during this period will fail, but the mail should queue up on the sending computer for a subsequent resend attempt.

    • .org vs .net
      2011 Mar. 28: 02:32. We have put redirects from many of the /iau HTML pages on the .org site to funnel users to the .net site. This is a preliminary step to the shutting down of the physial .org site.

    • Old Epoch Elements for Numbered Periodic Comets in Comet Element Files
      2011 Mar. 24: 13:20. An observer has pointed out a problem with the comet element files accessible in the Ephemerides and Orbital Elements service. We have determined that the problem is with the epoch of some of the orbits. We are investigating and will regenerate the files when the solution is found.
      • 14:50. Corrected files have been put in place.

    • Retirement of Ubasti Web Server (April 18)
      2011 Mar. 23: 15.25. The http server on the old web server machine (the old destination for the alias 'scully') will be shutoff sometime on Monday, 18 April. This date is not open to change as it is dependent on changes in the CfA's internal network that will be implemented at the same time. This unchangeable (by us!) date has now been changed twice, once to allow more testing of the internal-network changes and once because of the uncertainty of whether the U.S. Government would be shut down.

      Most cgi scripts are now running on the new script server. The status of porting the cgi scripts is detailed here.

    • Date of Last Update to NEOCP Not Updating on Main Page
      2011 Mar. 16: 14:00. It has been brought to our attention that the date of last update to the NEOCP that is present on the main MPC observers page (/iau/mpc.html) has not been updated since Feb. 28.54 UT. We are investigating.
      • 20:40. The problem has been located and, we believe, fixed.

    • AUTOACK Problem Overnight
      2011 Mar. 12: 18:00. Yesterday evening, the AUTOACK process started failing. No ACKs were being sent out. No observer brought this to our attention until early this morning. We have located the problem, fixed it and have resubmitted the affected batches to AUTOACK. If you use the AC2 keyword in your headers, you should receive the ACK. If you don't use the AC2 keyword, you will not.

    • .org vs .net
      2011 Feb. 22: 22:00. Over the next month we will begin to cease updating certain pages on the .org site, in anticipation of removing the physical site in the near future. The .org site was set up as an emergency host following the events of last February. The hosting site has a number of problems, principally the throttling of transfer rates for large files, that cause us particular grief in trying to maintain both the .net and .org sites. The first pages to be updated only on the .net site will be the ECS pages associated with mid-month MPS and monthly MPC batches. These .org pages will be modified to redirect automatically to the .net site. When the physical .org site is removed, it will be set up to redirect to .net.

    • Problem With Certain Objects in MPES
      2011 Jan. 31: 15:20. There seems to be a problem with the return of ephemerides of certain recently-designated objects in the MPES. A request for an ephemeris just sits at the "waiting for a response" message. Other objects are unaffected and problems (based on personal experience) appear to occur only with certain hosts. At this stage, we do not understand why the problem is occurring.
      • Feb. 1: 01:00. We discovered that the problem was due to incorrect protections on the variant orbit files. The immediate problem was fixed and the nightly update procedure was modified to set the correct protections.

    • The MPC Web Site
      2011 Jan. 19: 15:30. The main MPC web site is now https://www.minorplanetcenter.net/. This link will take you to the new interface, with access to new features such as the MPC Database. For those who are used to the earlier incarnation of the site, the former page layout is accessible at https://www.minorplanetcenter.net/iau/mpc.html. The ".org" site will, at some point in the near future, redirect users to the .net site.

    • Move over to new script server II
      2011 Jan. 5: 16:00. Most of the reported problems with the changeover have been fixed. Variant orbits are currently being copied, to be followed by residual blocks, for use in the MPES. The problem with incorrect elements being returned for certain unnumbered minor planet designations is believed to be fixed.

    • Move over to new script server
      2011 Jan. 4: 17:15. We have made a request to the CfA's networking group to shift the machine name "scully" (which is an alias) to the new script server machine. Some scripts will continue to run on the old server machine for sometime longer. It is to be expected that quite a number of links will need fixing. Please use the feedback forms to report problems. We had hoped that the change over would be completely transparent to outside users, but the change in the webserver from Purveyor to Apache makes this impossible. Some notes on the changes are available and contains details on how to modify any user-written scripts that call the MPC scripts directly.

    • Certain numbered orbits in MPES
      2011 Jan. 3: 15:23. Due to a repeated set of elements for a numbered minor planet, the MPES was returning the incorrect orbit for requests for numbered minor planets with numbers above (217439). That incorrect misbehavior has been fixed. A problem remains in that the name or principal desigantion associated with the numbered object is being displayed incorrectly.