[nos-bbs] JNOS error tracking new messages

Michael Fox - N6MEF n6mef at mefox.org
Thu Sep 13 04:10:01 EDT 2018

I think I discovered a problem with how JNOS tracks new messages.  If you're
familiar with the spool/mail/xxx.usr file and how JNOS tracks which messages
have been seen, please let me know if I've got this right.

When a user changes to an area/mailbox, JNOS summarizes how many messages
are there, including how many are new.


A foo
foo: 15 messages  -  4 new

The number of new messages appears to be determined by looking at how many
messages have ID numbers greater than the number stored next to the user's
call sign in the spool/mail/foo.usr file.


If foo.usr contains:

n6mef 12345

Then, the last time n6mef logged in, the message in foo.txt with ID AA12345
was the last message listed by the user.

Since this ID number doesn't grow without bound, it must "roll over" when it
reaches it's maximum value.  Since the ID numbers never seem to grow larger
than 5 digits, I'm assuming that they are 16 bit numbers.  So when the
number gets up to 65535, it must role over to 0.

Now, here's the problem.

We have seen many cases where a user switches to a mailbox and sees "X
messages - Y new", where Y is not correct.  For example, maybe it says "...
0 new" when, in fact, there are messages in that mailbox that have arrived
since the last time the user logged in.  In each of those cases, the ID
number next to a user's call sign in the foo.usr file is larger than the IDs
of the new messages in the foo.txt file. 

My assumption is the ID numbers have rolled over since the last time the
user checked the mailbox and the new messages have lower numbers than what
is stored in foo.usr.
So, JNOS tells the user "0 new", even though there are new messages that
user hasn't seen.  And if this assumption is correct, it seems to me that
this can never work since there will always be cases where the number rolls
over after a user has last checked the mailbox.

Is my understanding correct?
If not, please describe how the "... X new" is determined.

Thanks in advance.


More information about the nos-bbs mailing list