home *** CD-ROM | disk | FTP | other *** search
- Error Messages for Citadel-86
- V3.32
- 90Jul08
-
- This is an incomplete listing of error messages, both fatal and warning,
- Citadel-86 can generate, along with possible explanations and even plans
- for fixing same. These can occur both in CONFG and CTDL.
-
- FATAL MESSAGES:
- These messages will usually show up at the console and in a file named
- CRASH.
-
- ------------
-
- "INTERNAL: No room for new option!"
-
- PROBLEM: This error indicates you have either a trashed CTDLPROT.SYS file
- or have defined either too many external upload or download protocols. The
- limit is 15 for both modes.
-
- SOLUTIONS: It should be obvious if the problem is in uploads or downloads.
- Remember, you're only allowed about 20 external protocols in each of uploads
- and downloads (not total, but per). Check to make sure your CTDLPROT.SYS is
- correct (not trashed up) -- if Ease thinks it's ok, then check it by hand.
- You'll find CTDLPROT.SYS in your #ROOMAREA directory.
-
- ------------
-
- "?getLog-read fail//EOF detected (1)!"
- "?getLog-read fail//EOF detected (2)!"
- "?getLog-read fail//EOF detected (3)!"
-
- PROBLEM: An attempt to read an account from the log failed. 1 indicates
- the failure occurred while reading the main body of the account, 2 the failure
- occurred while trying to read the room pointers, 3 the failure occurred in
- reading the Mail> data. This is a serious problem.
-
- SOLUTIONS? If the cause isn't a damaged CTDLLOG.SYS which will require you
- to erase it and reconfigure (and thus lose all of your accounts if you haven't
- backed up), then there are a few other causes:
-
- o Changing #MAIL-SLOTS in CTDLCNFG.SYS without running one of the data
- expansion contraction utilities (Ease is preferred, but the obsolete DataChng
- also will do the job).
-
- o Changing #MAXROOMS in CTDLCNFG.SYS without etc.
-
- o Changing both #MAXROOMS and #MAIL-SLOTS in the same session with Ease.
- While this should be theoretically possible, it is NOT recommended.
-
- o Internal bug. Report it if the other causes seem improbable.
-
- ------------
-
- "?putLog-write fail (1)!"
- "?putLog-write fail (2)!"
- "?putLog-write fail (3)!"
-
- PROBLEM: As you might guess, these are related to those above. They indicate
- an attempt to update the log failed. 1 indicates the failure occurred
- while writing the main body of the account, 2 while writing the room pointers,
- and 3 the Mail> data. This problem is very serious. The most probable
- cause is some sort of odd mixup in MS-DOS, or an internal bug.
-
- ------------
-
- "?getRoom(): read failed//error or EOF (1)!"
- "?getRoom(): read failed//error or EOF (2)!"
-
- PROBLEM: An attempt to read a room record failed. 1 indicates the failure
- came while reading the main body of the room, 2 the messages themselves.
-
- SOLUTIONS? This is a serious problem. It can be caused by changing
- #MAXROOMS without employing Ease or DataChng. Otherwise, you may have found a
- bug.
-
- ------------
-
- "?putRoom(): fseek failure!"
- "?putRoom() crash!//0 returned!!!(1)"
- "?putRoom() crash!//0 returned!!!(2)"
-
- PROBLEM: A room record couldn't be written; 1 indicates the failure came
- while writing the main body of the room's record, 2 the messages. The fseek
- failure indicates the system wasn't allowed to find the correct place in
- CTDLROOM.SYS to write.
-
- SOLUTIONS? This is very serious and may indicate a DOS problem or internal
- bug.
-
- ------------
-
- "Memory failure -- I need more memory!"
-
- PROBLEM: Citadel-86 couldn't allocate memory for an internal buffer.
-
- SOLUTION: Don't run such a large system or free up more space by killing
- a TSR or shrinking a RAM disk.
-
- ------------
-
- "Couldn't open the floor file for update!"
-
- PROBLEM: Citadel-86 tried to open the floor file for an update and failed.
-
- SOLUTION: This may be caused by accidentally deleting your CTDLFLR.SYS
- file while in Outside Commands or something. Or it may indicate you don't
- have FILES=20 in your CONFIG.SYS.
-
- ------------
-
- "?putFloor(): write failed!"
-
- PROBLEM: Citadel-86 tried to write to the floor file and failed.
-
- SOLUTION? The only probable cause would be a problem with disk full or
- with DOS.
-
- ------------
-
- "?getNet-read fail(1)!!"
- "?getNet-read fail(2)!!"
- "?getNet-read fail(3)!!"
-
- PROBLEM: An attempt to read a net (node) record failed. 1 indicates the
- failure occurred while reading the main body of the net record, 2 the room
- sharing information, 3 the network-archive information.
-
- SOLUTION: The causes could be
-
- o changing #SHARED-ROOMS without using Ease or DataChng.
- o changing #NET-ARCH-ROOMS etc.
- o trying to change both of the above at the same time.
- o internal bug.
-
- ------------
-
- "?putNet-write fail(1)!!"
- "?putNet-write fail(2)!!"
-
- PROBLEM: An attempt to write a net (node) record failed. 1 indicates the
- failure occurred while writing the main body of the net record, 2 the room
- sharing information.
-
- SOLUTIONS? Well, best bets are a full disk, an internal bug or a problem
- with DOS.
-
- ------------
-
- "?nextMsgChar-read fail"
-
- PROBLEM: Citadel-86 tried to read a sector from the message base and
- failed.
-
- SOLUTIONS? Either you have a badly damaged CTDLMSG.SYS (the file has
- been truncated in some way), or you've messaged with your #MESSAGEK parameter
- without using Expand or Ease.
-
- ------------
-
- "?startAt read fail"
-
- PROBLEM: Citadel-86 tried to change file position to read a message and
- wasn't allowed to.
-
- SOLUTIONS? See above error.
-
- ------------
-
- "?putMsgChar-write fail"
- "?putMsgChar-read fail"
- "?ctdlmsg.sys write fail"
-
- PROBLEM: These two errors indicate, collectively, a failure to write a
- message to disk. The first and third indicate a failure occurred while
- actually writing to disk, the second failure occurred while reading in part
- of the message base in preparation for modification and writing.
-
- SOLUTIONS? See above error.
-
- ------------
-
- "putMessage -- couldn't open direct mail file!"
- "putMLNet crash"
-
- PROBLEM: Citadel-86 attempted to open a network Mail> file (*.ML) and
- failed; Citadel-86 couldn't write to the network Mail> file.
-
- SOLUTIONS? The disk may be full; #NETAREA may have disappeared (unlikely,
- though); internal bug.
-
- ------------
-
- "putMessage -- couldn't open route mail file!"
-
- PROBLEM: Citadel-86 attempted to open a network routed Mail> file (R*.*)
- and failed.
-
- SOLUTIONS? See above error.
-
- ------------
-
- "bad op: start=<number> cycles=<number>"
-
- PROBLEM: An internal Citadel-86 routine contained a bad op-code.
-
- SOLUTIONS? Memory is being corrupted by something. Try a new version of
- Citadel-86, see if you can get it to reproduce (and then report it), try
- changing something in your system.
-
- ------------
-
- "Dependent variables mismatch!"
-
- PROBLEM: Citadel-86 detected a a mismatch between the room name in the
- internal index and the name of the room read off of disk.
-
- SOLUTIONS? Internal bug. Report it, see if you can reproduce it at will.
-
- ------------
-
- "shared rooms: #2"
- "shared rooms: #1"
-
- PROBLEM: A bad shared room status was detected.
-
- SOLUTIONS? An examination of the netlog.sys should indicate what room the
- crash occurred for; also, editing the implicated node and using the 'R'ooms
- command should show which room looks odd. Unshare the room and then reshare
- it.
-
- ------------
-
- "ctdlvnet.sys is missing!!"
- "ctdlvrm.sys is missing!"
- "ctdlvnet.sys is missing!!"
-
- PROBLEM: These indicate something's wrong in the virtual rooms.
-
- SOLUTIONS? Check to make sure FILES=20 is in your CONFIG.SYS. See if
- the files exist in the VIRTUAL directory.
-
- ------------
-
- "crashout: unknown case in switch statement"
-
- PROBLEM: Citadel-86 is upset about handling the calllog.sys.
-
- SOLUTIONS? This is an internal bug, please report it and all circumstances.
-
- ------------
-
- "CallMsg error!"
-
- PROBLEM: Citadel-86 can't open an audit file.
-
- SOLUTIONS?
-
- o The most likely problem is FILES=20 is not in your CONFIG.SYS. Triple
- check this possibility. Or maybe you didn't reboot after setting it in your
- CONFIG.SYS?
- o Citadel-86 "moved" itself out of its directory and failed to move back (very
- unlikely), which is an internal bug.
-
- ------------
-
- "putSLNet crash"
-
- PROBLEM: Citadel-86 can't write to a sendfile file.
-
- SOLUTIONS? Beats us.
-
-
- ------------
-
- NON-FATAL WARNING MESSAGES:
- These messages typically only show up either at the console or on the
- user's screen and the console.
-
- ------------
-
- "URP IN DOMAINS!"
-
- PROBLEM: Internal bug in Citadel-86. Indicates the system thought was
- absolutely certain it is a domain server for some domain, but upon checking
- discovered to its dismay that it is not.
-
- SOLUTIONS? Report it and all details. Attempt to reproduce.
-
- ------------
-
- "WARNING: badly formed ctdldmhd.sys."
-
- PROBLEM: Your CtdlDmhd.Sys file (see Network3.Man) has one or more lines
- in it that Citadel-86 couldn't really do much with.
-
- SOLUTIONS? Fix them if you built the file yourself, else contact the
- person who made the file and inform them of the problem. Be aware that maybe
- the file has just been trashed up. Also, if you have a CtdlDmhd.Lcl file
- in place, the problem may lie in there rather than in CtdlDmhd.Sys.
-
- ------------
-
- "INTERNAL FILE ERROR!"
-
- PROBLEM: Citadel-86 tried to open a compressed file prepatory to reading
- information from it and failed.
-
- SOLUTIONS? At this point, Citadel-86 should already have verified the file
- exists, so why the opening the file failed indicates a bug of some sort.
-
- ------------
-
- "Sorry, no room in system."
-
- This message is displayed to a TWITted user attempting to create a room.
-
- ------------
-
- "Your date is unintelligible."
-
- This message is displayed to a user who attempted to read messages since
- or before a given date and entered a date Citadel-86 couldn't decode. The
- correct format for a date is yymmmdd, where yy is optional, mmm is the
- month and is text, not numeric, and need only be so long as to be unique
- (the longest this requires is three letters in the case of July and June).
-
- ------------
-
- "System error with mkdir!"
-
- This message is displayed when attempting to create a temporary work
- directory. The only probable cause is a full disk, since Citadel-86 checks
- for files/directories which might interfere with creating a temporary
- directory (i.e., the name used for the temporary directory is dynamic).
-
- ------------
-
- "SetSpace failure, aborting!"
-
- This message indicates the system attempted to change directory to the
- directory attached to the current room prepatory to transmitting a file
- using an external protocol and failed. Presumably, the existence of the
- directory should have been verified, so this error would probably indicate
- a bug of some sort.
-
- ------------
-
- "Sorry, the list of files would be too long. Try again without a date
- specification."
- "Sorry, the list of files would be too long."
-
- These two messages are caused by the same problem. Citadel-86 transmits
- files via external protocols by executing the named program with the file
- names (wildcards are expanded). Unfortunately, DOS has a command line
- limitation of 128 characters, so if too many file names result from the
- user's file specification, Citadel-86 cannot run the program safely. If there
- is a date specification attached to the download, then Citadel-86 simply
- aborts. If not, Citadel-86 again tries to build a command line, this time
- with an unexpanded file-spec (i.e., what the user typed in). If overflow
- once again occurs, Citadel-86 gives up. If it does work, Citadel-86 will
- execute the external program, trusting the program can expand its wildcards.
-
- ------------
-
- "Ooops, couldn't find <msg#>"
-
- This message indicates your system's message base (CTDLMSG.SYS) has
- suffered some sort of damage, or Citadel-86 is confused. Running CONFG
- on your system may fix it if Citadel-86 is confused (although you may have
- to delete CHKPT to ensure a complete reconfigure), but if there's real
- damage (caused by disk errors or Citadel-86 bugs (the latter unlikely)),
- then you'll just have to wait until Citadel-86 overwrites the damaged error.
-
- Since we've not seen Citadel-86 destroy its own message base, you may want
- to consider have your hardware checked.
-
- ------------
-
- "Unexpected internal error, please report it!"
-
- This message indicates a bug in Citadel-86's Mail system. Please try
- to reproduce it and then report it.
-
- ------------
-
- "Internal error in mail!"
-
- This error indicates another bug in Citadel-86's Mail which should
- be reported. This time, Citadel-86 was saving a Mail message and suddenly
- discovered the user the Mail was intended for can no longer be found
- despite earlier verification!
-
- ------------
-
- "Internal error, couldn't identify <string>"
-
- This error, which can involve both users and system names, is generated
- by the Who Else processing and, once again, indicates an internal bug
- which should be reproduced and then reported it.
-
- ------------
-
- "ERROR: Couldn't open output file <filename>"
-
- This message is occurs when Citadel-86 attempted to open a file to Journal
- (manual) or Archive (automatic) a message and the open failed. This
- normally indicates a directory or other special object in DOS already
- exists with that name, or you tried to open a file which can't exist
- (usually because the directory you specified also doesn't exist). Please
- use a different name.
-
- ------------
-
- "C-86 internal error with doors. Sorry."
-
- Citadel-86 tried to open a transitional data file (to save data in while
- a door is up) and failed. Maybe your disk is full? Maybe we have a bug
- in Citadel-86.
-
- ------------
-
- "?ERROR CREATING!"
-
- This message occurs when Citadel-86 tries to create a directory for some
- reason (usually during System Operator activity). The only probable cause,
- offhand, is a full disk.
-
- ------------
-
- "That's not a directory!"
-
- Citadel-86 asked you for a directory name and you gave it the name of
- a file or something else which it couldn't process. BAAAAAAAAAAAAAAAD
- Sysop!
-
- ------------
-
- "?Directory not present! (internal error)"
-
- This message indicates your CTDLDIR.SYS file is messed up or missing
- completely. Check on it.
-
- ------------
-
- "?Directory not present!"
-
- This message indicates the directory attached to the current room
- doesn't exist. You probably deleted it while Citadel-86 was off-line.
- Either recreate it or change the room's setting.
-
- ------------
-
- "Unknown problem with filetagging!"
-
- FILEDIR.TXT could not be opened for update. Disk full? Bug?
-
- ------------
-
- "buffer overflow"
-
- The user tried to type in more than 7500 characters of text. Split the
- message.
-
- ------------
-
- "?not found."
-
- User using replace string typed in a string which isn't in the current
- message.
-
- ------------
-
- "?Overflow!"
-
- User using replace string typed in a replacement string which, if allowed,
- would cause an overflow of the message buffer. Split the message.
-
- ------------
-
- "Sorry, already sharing <number> rooms with <nodename>"
-
- You've reached your #SHARED-ROOMS setting for the named node. Either
- give up or use Ease or DataChng to increase #SHARED-ROOMS.
-
- ------------
-
- "Couldn't open %s for update?"
-
- This happens when you are attempting to send some files to another system
- and Citadel-86 couldn't open a file to handle the administration. Full disk,
- perhaps? Bug?
-
- ------------
-
- "Sorry, reserved name"
-
- You typed in &L for a nodename. This is reserved to indicate "All Local
- Systems."
-
- ------------
-
- "There are only 31 nets to choose from."
-
- Read NETWORK3.MAN on Member Nets.
-
- ------------
-
- "Couldn't append to <filename>????"
-
- You wanted to request some files but this error occurred. Full disk?
- Bug?
-
- ------------
-
- "Unknown problems with netMail: author=-<name>-, target=-<name>-,
- context=-<stuff>-. System was <name>."
-
- This error, which will show up in the Aide> room, indicates your system
- attempted to send Net Mail to another system (named in the message), but
- failed for some completely unknown reason. The author name indicates who
- authored the message, the target is the name of the person the mail was
- supposed to be delivered to, and the context is supposed to contain stuff
- allowing debug of the problem.
-
- The only current case in which this may show up is when someone sends
- netmail somewhere else but to a recipient who doesn't exist, and then is
- deleted from the sending system before the mail is delivered. This can also
- happen on older systems (pre-V3.32) with accounts forwarding mail incorrectly
- onward, when the mail being forwarded is netmail and the sender on real
- originating system doesn't happen to exist on the intermediate system (that
- is, the one which found the warning message in the Aide> room).
-
- ------------
-
- THE NETLOG
-
- These messages may appear from time to time in your netlog if you happen
- to run a netlog. Not all netlog messages are covered here, only the unusual
- messages.
-
- ------------
-
- "Couldn't switch to WXMODEM"
- "No YMODEM"
-
- These two messages indicate Citadel-86 couldn't switch from XMODEM, the
- default low-level protocol, to a different (and presumably better)
- protocol. These messages will be seen most often when dealing with
- older Citadel-86 variants such as STadel or very old Citadel-86 systems.
-
- ------------
-
- "No compaction"
-
- For some reason Citadel-86 couldn't enable message compaction when netting
- with another system. As with the two previous messages, this'll be seen
- with old variants and elderly Citadel-86 systems.
-
- ------------
-
- "ITL_send failure <number>, mode <number>"
-
- This message indicates an unexpected interruption of service at the
- lowest level of the network protocol, specifically during transmission (as
- opposed to reception) of data or control information. There are a number of
- causes for messages of this sort:
-
- o Some versions of STadel and STadel-variants do not handle role-reversal
- termination correctly. In this case there is nothing to worry about, all
- data should transfer cleanly despite the unfortunate glitch.
-
- o Line noise. There is no guarantee that all data transferred.
-
- o Disk full conditions sometimes cause this.
-
- o Etc.
-
- The following are the specific values and their meanings for the two
- numbers that can show up.
-
- The FAILURE code (first number):
-
- 2 -- The receiver cancelled the transmission. This is highly unusual,
- since a cancellation of a transmission should take place at a far higher
- level.
-
- 3 -- The transmission never even started; the initialization sequence for
- whatever protocol is in use (NAK for XMODEM, for example) was never
- received during the timeout period.
-
- 4 -- General failure code, could mean anything from the system was being
- fussy to full disks.
-
- 9 -- Carrier was lost.
-
- The MODE code (second number) can only be one of two numbers:
-
- 1 -- The failure occurred while attempting to start a transmission; that is,
- Citadel-86 was prepared to send information and the other end crapped out
- without even trying to initialize the transmission. Usually, Mode 1 failures
- occur with a failure code of 9.
-
- 2 -- The failure occurred during or at the end of the transmission.
-
- ------------
-
- "Msg from unknown node <system name> @<system id>",
-
- During vortex handling, Citadel-86 displays the name and ID of every
- system your installation doesn't know about which had a message in
- the current batch to be processed.
-
- ------------
-
- "<message id> from <system name> rejected."
-
- This indicates a message is thought to have been vortexed and was
- thus rejected. The message may be in a file named DISCARD.
-
- ------------
-
- "Couldn't create <filename>!!"
-
- Citadel-86 was going to create a vortex record for some system and
- couldn't open the file for recording the data. Is your disk full?
-
- ------------
-
- "No recipient: <name>"
-
- Net Mail> was sent to a non-existent user on your system.
-
- ------------
-
- "WARNING: Can't open <filename>!!!!"
-
- Citadel-86 couldn't open a temporary buffer for generating the summary
- Aide> message. Disk full? FILES=20 in your CONFIG.SYS?
-
- ------------
-
- "Rejecting <system name> (<system id>)"
-
- The other system wanted to use you to route to another system and
- you refused for some reason, most likely route flags.
-
- ------------
-
- THE DOMAIN.LOG
-
- These messages may appear from time to time in your domain.log (which in
- turn is located in your #DOMAINAREA directory).
-
- ------------
-
- "Mail for unknown system <system name>."
-
- PROBLEM: Your system is a domain server. Your system was asked to deliver
- mail to a system in a domain you serve which you couldn't identify.
-
- SOLUTIONS? If the system really is unknown, not much. You can delete the
- mail if you like; mail addresses are located in the first 42 bytes of each
- file, and killing a file won't hurt. At the current time there is no mechanism
- for automatically sending negative acknowledgements back to the sender.
-
- If, on the other hand, the system name was merely misspelled, you can use
- your ALIASES.SYS file to redirect the mail. See Network3.Man for more detail
- on ALIASES.SYS.
-
- Finally, if it happens that the target system is located in some other domain
- which you do not serve (user picked wrong domain), you can use an editor to
- edit your MAP.SYS file to redirect the mail. Simply find the entry containing
- the bad domain name, correct it, and bring your system back up (oh yeah, you
- should bring your system down before doing this sort of stuff).
-
- ------------
-
- "Don't know how to reach domain <domain name>."
-
- PROBLEM: This is a very rare message. Your system has no mail hub defined
- and has received mail for a domain it knows nothing about.
-
- SOLUTIONS? Define a mail hub. Find out who serves the domain and make
- appropriate adjustments to your system to get it delivered. If the domain
- is just misspelled, consult the last paragraph of the above error message.
-
- ------------
-