Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.sys.acorn.networking > #5174 > unrolled thread

Hermes logs

Back to article view | Back to comp.sys.acorn.networking

Contents

  Hermes logs Richard Porter <ricp@minijem.plus.com> - 2024-10-04 11:22 +0100
    Re: Hermes logs Paul Sprangers <Paul@sprie.nl> - 2024-10-04 13:02 +0200
      Re: Hermes logs Chris Hughes <news13@noonehere.co.uk> - 2024-10-04 13:31 +0100
    Re: Hermes logs Chris Hughes <news13@noonehere.co.uk> - 2024-10-04 12:04 +0100
      Re: Hermes logs Steve Fryatt <news@stevefryatt.org.uk> - 2024-10-05 13:06 +0100
    Re: Hermes logs druck <news@druck.org.uk> - 2024-10-05 02:11 +0100

#5174 — Hermes logs

I've just had a look at the log for one Send account. It contains normal 
interactions up to 20 Sep., followed by "Failed to connect..." messages up 
to:
[30 Sep 2024 22:41:52] Connecting to <server>
[00][00][00][00][00][00][00][00][00]...

714,745 nulls. Yes that's ~0.7 million nulls in all!

...[00][00][00][00][00[03 Oct 2024 14:50:05] Failed to connect to <server>

Then it went back to "Connecting to <server>"
"Failed to connect to <server>".

All I can say is WTF? Would the nulls have come from the server? I haven't 
seen this on any other log.

-- 
Richard Porter

[toc] | [next] | [standalone]

#5175

In article <d91d81aa5b.news@user.minijem.plus.com>,
   Richard Porter <ricp@minijem.plus.com> wrote:

> [...]
> [00][00][00][00][00][00][00][00][00]...

Although I'm familiar with mysteriously added [00]'s to the end of
different types of files, the Hermes logs on my system appear to be clean.

It does rise another question though: these log files occupy 220+ MByte of
disk space. Is it safe to delete them?

Paul

-- 
https://riscos.sprie.nl

[toc] | [prev] | [next] | [standalone]

#5177

In message <5baa84d9cePaul@sprie.nl>
          Paul Sprangers <Paul@sprie.nl> wrote:

> In article <d91d81aa5b.news@user.minijem.plus.com>,
>    Richard Porter <ricp@minijem.plus.com> wrote:

>> [...]
>> [00][00][00][00][00][00][00][00][00]...

> Although I'm familiar with mysteriously added [00]'s to the end of
> different types of files, the Hermes logs on my system appear to be clean.

> It does rise another question though: these log files occupy 220+ MByte of
> disk space. Is it safe to delete them?

Yes, they just carry on after you empty them. But there is a TRIM logs 
button to Trim the files on the Hermes Accounts windows - mine are set for 
40 kb but you can change the figure to whatever you want.



-- 
Chris Hughes

[toc] | [prev] | [next] | [standalone]

#5176

In message <d91d81aa5b.news@user.minijem.plus.com>
          Richard Porter <ricp@minijem.plus.com> wrote:

> I've just had a look at the log for one Send account. It contains normal
> interactions up to 20 Sep., followed by "Failed to connect..." messages up
> to:
> [30 Sep 2024 22:41:52] Connecting to <server>
> [00][00][00][00][00][00][00][00][00]...

> 714,745 nulls. Yes that's ~0.7 million nulls in all!

> ...[00][00][00][00][00[03 Oct 2024 14:50:05] Failed to connect to <server>

> Then it went back to "Connecting to <server>"
> "Failed to connect to <server>".

> All I can say is WTF? Would the nulls have come from the server? I haven't
> seen this on any other log.

Very likely it will be recording what the server is sending in resposne to 
your request to connect.

-- 
Chris Hughes

[toc] | [prev] | [next] | [standalone]

#5179

On 4 Oct, Chris Hughes wrote in message
    <03fd84aa5b.chris@mytardis>:

> Very likely it will be recording what the server is sending in resposne to
> your request to connect.

On RISC OS, it's at least equally likely that the open log files have been
correpted by another process, as Druck says.

-- 
Steve Fryatt - Leeds, England

http://www.stevefryatt.org.uk/

[toc] | [prev] | [next] | [standalone]

#5178

On 04/10/2024 11:22, Richard Porter wrote:
> I've just had a look at the log for one Send account. It contains normal
> interactions up to 20 Sep., followed by "Failed to connect..." messages up
> to:
> [30 Sep 2024 22:41:52] Connecting to <server>
> [00][00][00][00][00][00][00][00][00]...
> 
> 714,745 nulls. Yes that's ~0.7 million nulls in all!
> 
> ...[00][00][00][00][00[03 Oct 2024 14:50:05] Failed to connect to <server>
> 
> Then it went back to "Connecting to <server>"
> "Failed to connect to <server>".
> 
> All I can say is WTF? Would the nulls have come from the server? I haven't
> seen this on any other log.

Log corruptions are normally one of two things, either a crash causing a 
write of a partially filled buffer, or another application starts 
writing something to the open log file handle, as RISC OS has no 
protection against that sort of ting.

---druck

[toc] | [prev] | [standalone]

Back to top | Article view | comp.sys.acorn.networking

csiph-web