[Dxspider-support] DXSpider stop responding to connections with 100% CPU

[Normando IZ6FXS]

Happened twice today. Anybody can help?
Can I activate more debug? Which?

Thanks,
Norm IZ6FXS

> Il giorno 19 feb 2024, alle ore 09:55, iz6fxs--- via Dxspider-support <dxspider-support at tobit.co.uk> ha scritto:
> 
> Hi!
>  
> Please help me troubleshooting this problem. DXspider (cluster.iz6fxs.radio) suddenly stops responding with the “login” prompt if you connect to it. I’m running the las build.
> When this happens the CPU is stuck to 100%:
>  
> <image003.jpg>
>  
> This is the content of the log:
>  
> root at cluster:/spider/local_data/log# tail 2024/02.dat 
> 1708206540^DXProt^PC92A E77AR -> 31.223.135.216 on DB0SUE-7
> 1708206545^DXProt^PC92A EB1FEV -> 88.10.193.136 on EA4URE-3
> 1708206551^DXProt^PC92A JJ3FBS -> 157.14.219.178 on JE3YEK
> 1708206551^DXProt^PC92A 4X6TT -> 147.235.199.58 on NX9G
> 1708206551^DXProt^PC92A W0FK -> 47.24.152.2 on NX9G
> 1708206552^DXProt^PC92A EA1AOC -> 91.196.223.124 on EA4URE-5
> 1708206561^DXProt^PC92A K4FTV -> 35.137.52.111 on EI7MRE
> 1708206562^DXProt^PC92A KI0EB -> 24.245.245.113 on W1NR
> 1708206567^DXProt^PC92A OK1CF -> 77.237.128.209 on EA4RCH-5
> 1708206568^DXProt^PC92A DF1MM -> 176.1.242.227 on S50CLX
>  
> And this is the contect of the last lines of the debug log:
>  
> root at cluster:/spider/local_data/debug# tail -20 2024/048.dat 
> 1708206345^(*) RBN:WRITE_CACHE size: 420.774KB time to write: 21 mS
> 1708206368^(*) RBN: ERROR invalid prefix/callsign T9CT from WB6BEE-# on 28005.9, dumped
> 1708206386^(nologchan) PC61^7012.0^W4NF^17-Feb-2024^2146Z^arrl^PA2A^SR2PUT^77.171.80.188^H26^~
> 1708206386^(*) PCPROT: Bad Spotter PA2A, dropped
> 1708206386^(nologchan) PC61^7012.0^W4NF^17-Feb-2024^2146Z^arrl^PA2A^SR2PUT^77.171.80.188^H23^~
> 1708206386^(*) PCPROT: Bad Spotter PA2A, dropped
> 1708206386^(nologchan) PC61^7012.0^W4NF^17-Feb-2024^2146Z^arrl^PA2A^SR2PUT^77.171.80.188^H24^~
> 1708206386^(*) PCPROT: Bad Spotter PA2A, dropped
> 1708206386^(nologchan) PC61^7012.0^W4NF^17-Feb-2024^2146Z^arrl^PA2A^SR2PUT^77.171.80.188^H23^~
> 1708206386^(*) PCPROT: Bad Spotter PA2A, dropped
> 1708206405^(*) RBN:WRITE_CACHE size: 420.606KB time to write: 18 mS
> 1708206406^(*) RBN: ERROR invalid prefix/callsign 1E2OCV from JN1ILK-# on 7007.5, dumped
> 1708206420^(err) SK0MMR connected from 44.52.120.88
> 1708206420^(*) RBN: noinrush: 0, setting inrushpreventor on SK0MMR to 0
> 1708206465^(err) RBN: no input from SK0MMR, disconnecting
> 1708206465^(*) RBN:WRITE_CACHE size: 424.195KB time to write: 22 mS
> 1708206480^(err) SK0MMR connected from 44.52.120.88
> 1708206480^(*) RBN: noinrush: 0, setting inrushpreventor on SK0MMR to 0
> 1708206525^(err) RBN: no input from SK0MMR, disconnecting
> 1708206525^(*) RBN:WRITE_CACHE size: 421.602KB time to write: 20 mS
>  
> I noticed that all the connections to the node are hung in CLOSE_WAIT (hundreds if not more):
>  
> <image002.jpg>
>  
> I rebuilt the file users.v3j to no avail.
> Restarting it with systemd is not working, you have to kill the process manually to have it restarted.
>  
> Please help! Thanks,
> Norm IZ6FXS
>  
> _______________________________________________
> Dxspider-support mailing list
> Dxspider-support at tobit.co.uk <mailto:Dxspider-support at tobit.co.uk>
> https://mailman.tobit.co.uk/mailman/listinfo/dxspider-support

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.tobit.co.uk/pipermail/dxspider-support/attachments/20240219/19ffa0dd/attachment-0001.htm>