• mis.log

    From Michael Pierce@1:340/201 to All on Sat Jun 20 12:11:30 2020
    anybody else have a problem with tailing mis.log file??
    raspbery pi4 (4gb) a46 build
    tail -f mystic/logs/mis.log or even cat mystic/logs/mis.log

    once it reaches a certain point (and it takes days to get there) the "log
    file" stops adding to the log.

    mis server still runs just fine, just not adding to the log file anymore?? shutting down mis and re-starting fixes the problem.

    next time it happens, I will capture the log file

    I run mis server in another console.

    Michael2
    telnet://mansion.dynv6.net:2323
    web://mansion.dynv6.net:1580

    --- Mystic BBS v1.12 A46 2020/05/17 (Raspberry Pi/32)
    * Origin: Mike's Mansion - Pdx,Or - mansion.dynv6.net:2323 (1:340/201)
  • From Deon George@3:633/509 to Michael Pierce on Sun Jun 21 10:18:07 2020
    Re: mis.log
    By: Michael Pierce to All on Sat Jun 20 2020 12:11 pm

    once it reaches a certain point (and it takes days to get there) the "log file" stops adding to the log.

    What's the size of it? Is it the same size each time?

    ...δεσ∩

    ... Half of conversation is listening.
    --- SBBSecho 3.11-Linux
    * Origin: I'm playing with ANSI+videotex - wanna play too? (3:633/509)
  • From Michael Pierce@1:340/201 to Deon George on Sat Jun 20 19:44:28 2020

    once it reaches a certain point (and it takes days to get there)
    the "log file" stops adding to the log.

    What's the size of it? Is it the same size each time?

    I will have to wait till it happens again.

    Michael

    ... Are we supposed to be having fun yet?
    --- GoldED+/LNX 1.1.5-b20180707
    * Origin: Mike's Mansion - Portland, OR - mansion.dynv6.net:2323 (1:340/201)
  • From g00r00@1:129/215 to Michael Pierce on Tue Jun 23 12:37:53 2020
    anybody else have a problem with tailing mis.log file??
    raspbery pi4 (4gb) a46 build
    tail -f mystic/logs/mis.log or even cat mystic/logs/mis.log

    once it reaches a certain point (and it takes days to get there) the "log file" stops adding to the log.

    MIS opens the log and keeps it open, only writing to the log file when it has 4 kilobytes of data to write. It does this to reduce wear and tear to the disk and it speeds things up a bit.

    You can disable that function by turning the log caching off. MIS will then update the log file immediate, but it has to open, write, and close the file each time a line needs to be added to the log. This is harder on your disk but you'll be able to tail it.

    You can turn it off in Servers > Logging > MIS.LOG > Cached Log > No

    --- Mystic BBS v1.12 A46 2020/06/11 (Windows/64)
    * Origin: Sector 7 | Mystic WHQ (1:129/215)
  • From Michael Pierce@1:340/201 to g00r00 on Tue Jun 23 16:10:27 2020
    anybody else have a problem with tailing mis.log file??
    raspbery pi4 (4gb) a46 build
    tail -f mystic/logs/mis.log or even cat mystic/logs/mis.log

    once it reaches a certain point (and it takes days to get there)
    the "log file" stops adding to the log.

    MIS opens the log and keeps it open, only writing to the log file when
    it has 4 kilobytes of data to write. It does this to reduce wear and
    tear to the disk and it speeds things up a bit.

    You can disable that function by turning the log caching off. MIS
    will then update the log file immediate, but it has to open, write,
    and close the file each time a line needs to be added to the log.
    This is harder on your disk but you'll be able to tail it.

    You can turn it off in Servers > Logging > MIS.LOG > Cached Log > No
    is there a way to FIX the caching so it doesn't die??

    don't want to decrease performance

    it happened again, log file hit 17megs, I copied the file to miscrash.log -rw-r--r-- 1 pi pi 17M Jun 23 16:07 miscrash.log

    I then shutdown mis, deleted mis.log and restarted

    the big question is why is it stopping writing??? and other than not caching, what else can be done ??? is there a way to keep track of mis via the console?

    Michael
    Raspberry Pi4 (4gb) usb-to-sata
    os raspberry os (32-bit)
    mem usuage 680M
    0 swap in use


    ... TV Truth #5: Drinking beer attracts beautiful females.
    --- GoldED+/LNX 1.1.5-b20180707
    * Origin: Mike's Mansion - Portland, OR - mansion.dynv6.net:2323 (1:340/201)
  • From g00r00@1:129/215 to Michael Pierce on Wed Jun 24 16:46:48 2020
    MIS opens the log and keeps it open, only writing to the log file whe it has 4 kilobytes of data to write. It does this to reduce wear and tear to the disk and it speeds things up a bit.

    is there a way to FIX the caching so it doesn't die??

    I don't think its dying it just doesn't write to the log until it gets enough data to write to the log, thats what caching is. So if you either want it to write in chunks of logs at a time, or you want it to write line by line.

    Maybe what I can do for MIS is have it update the log after a number of
    seconds even if the cache isn't filled yet.

    --- Mystic BBS v1.12 A46 2020/06/11 (Windows/64)
    * Origin: Sector 7 | Mystic WHQ (1:129/215)
  • From Michael Pierce@1:340/201 to g00r00 on Wed Jun 24 19:58:17 2020
    I don't think its dying it just doesn't write to the log until it gets enough data to write to the log, thats what caching is. So if you
    either want it to write in chunks of logs at a time, or you want it to write line by line.

    well it stops writing, but the mis server window keeps writing ie last time I caught it, there was more than 2 hours time difference between what was written to mis.log and what was on the mis server screen (running as a detached screen)

    poll.log does not have this problem.

    Maybe what I can do for MIS is have it update the log after a number
    of seconds even if the cache isn't filled yet.
    one thing I have noticed, is when it happens, it will only write the date

    here is the last few lines of mis from the last time it happened

    + 2020.06.24 19:50:54 BINKP > Connect on slot 1/5 (73.11.56.74)
    + 2020.06.24 19:50:54 BINKP 1-HostName c-73-11-56-74.hsd1.or.comcast.net
    + 2020.06.24 19:50:54 BINKP 1-Country United States of America (US)
    + 2020.06.24 19:50:54 BINKP 1-System fluxcap.dynv6.net
    + 2020.06.24 19:50:54 BINKP 1-SysOp Michael Pierce
    + 2020.06.24 19:50:54 BINKP 1-Location Portland, Or
    + 2020.06.24 19:50:54 BINKP 1-Mailer binkd/1.1a-96/Linux binkp/1.1
    + 2020.06.24 19:50:54 BINKP 1-Queued 0 files for 1:340/201.2@fidonet
    + 2020.06.24 19:50:54 BINKP 1-Queued 0 files for 21:2/145.2@fsxnet
    + 2020.06.24



    Michael

    ... Cap'n, I threw your stinkin palm tree overboard!
    --- GoldED+/LNX 1.1.5-b20180707
    * Origin: Mike's Mansion - Portland, OR - mansion.dynv6.net:2323 (1:340/201)