distribution that seems interesting (I am not sure if it will serve your
purpose) is SME server at
http://www.smeserver.org/
I have not tried it, but from what I have read it seems to have most of the
functionality of a server. It might not be exactly what you need but it
might be a good learning platform to see how things are done and set up.
Message
From: Derek Manson [mailto:dmanson (AT) avotive (DOT) com]
Sent: Thursday, January 25, 2007 9:30 AM
To: redhat-list (AT) redhat (DOT) com
Subject: Fedora 6 Advice
Good morning everyone,
I am seeking some advice on how to move forward with our plans to
integrate Fedora 6 into our network. We are a small town ISP currently
running in an all Windows network. Moving into this new year we are
discovering that our costs to run this type of network is getting very
costly. We are looking to move our entire services to Linux and preferably
Fedora. I have done a bit here and there with Fedora 5 and 6, however, not
to this large of scale. My question is how and where should I be looking
to get a really good understanding of how to operate in Fedora 6? We are
needing to setup a Fedora an email and web solution rather soon. I have
read a number of tutorials, forums and even some published books, I just
need a bit more. Any suggestions?
Thanks.
Derek M.

I am sure I am going to get a large, heavy rock thrown at me but in
some cases, I have found Ubuntu 6.06 Server to be an extremely solid and
very straightforward 'drop-in' server for certain types of roles;
www/ftp/smtp/mysql for example are incredibly easy to install, and the
documentation at the moment is very good. Hardware support is quite good,
and the installer is very intelligent. Their are cases where I think
CS/RHEL is better, but I have been let down by the redhat architecture a
few times lately.

katsu

1/26/07, Rokicki, Andrew (Andrew) <ARokicki (AT) ofsoptics (DOT) comwrote:

distribution that seems interesting (I am not sure if it will serve
your
purpose) is SME server at
http://www.smeserver.org/

I have not tried it, but from what I have read it seems to have most of
the
functionality of a server. It might not be exactly what you need but it
might be a good learning platform to see how things are done and set up.
--
Message
From: Derek Manson [mailto:dmanson (AT) avotive (DOT) com]
Sent: Thursday, January 25, 2007 9:30 AM
To: redhat-list (AT) redhat (DOT) com
Subject: Fedora 6 Advice

Good morning everyone,

I am seeking some advice on how to move forward with our plans to
integrate Fedora 6 into our network. We are a small town ISP currently
running in an all Windows network. Moving into this new year we are
discovering that our costs to run this type of network is getting very
costly. We are looking to move our entire services to Linux and preferably
Fedora. I have done a bit here and there with Fedora 5 and 6, however, not
to this large of scale. My question is how and where should I be looking
to get a really good understanding of how to operate in Fedora 6? We are
needing to setup a Fedora an email and web solution rather soon. I have
read a number of tutorials, forums and even some published books, I just
need a bit more. Any suggestions?

Thanks.
Derek M.

katsumi liquer wrote:
I am sure I am going to get a large, heavy rock thrown at me but
in
some cases, I have found Ubuntu 6.06 Server to be an extremely solid and
very straightforward 'drop-in' server for certain types of roles;
www/ftp/smtp/mysql for example are incredibly easy to install, and the
documentation at the moment is very good. Hardware support is quite good,
and the installer is very intelligent. Their are cases where I think
CS/RHEL is better, but I have been let down by the redhat
architecture a
few times lately.
No rock at all but tell us about letting down, I am interested when it
failed you, maybe we'll avoid such problems.
10ks

katsu

1/26/07, Rokicki, Andrew (Andrew) <ARokicki (AT) ofsoptics (DOT) comwrote:
>>
>distribution that seems interesting (I am not sure if it will serve
>your
>purpose) is SME server at
>http://www.smeserver.org/
>>
>I have not tried it, but from what I have read it seems to have most of
>the
>functionality of a server. It might not be exactly what you need but it
>might be a good learning platform to see how things are done and set up.
>>
>>
>Message
>From: Derek Manson [mailto:dmanson (AT) avotive (DOT) com]
>Sent: Thursday, January 25, 2007 9:30 AM
>To: redhat-list (AT) redhat (DOT) com
>Subject: Fedora 6 Advice
>>
>Good morning everyone,
>>
>I am seeking some advice on how to move forward with our plans to
>integrate Fedora 6 into our network. We are a small town ISP currently
>running in an all Windows network. Moving into this new year we are
>discovering that our costs to run this type of network is getting very
>costly. We are looking to move our entire services to Linux and
>preferably
>Fedora. I have done a bit here and there with Fedora 5 and 6,
>however, not
>to this large of scale. My question is how and where should I be looking
>to get a really good understanding of how to operate in Fedora 6? We are
>needing to setup a Fedora an email and web solution rather soon. I have
>read a number of tutorials, forums and even some published books, I just
>need a bit more. Any suggestions?
>>
>Thanks.
>Derek M.
>>
>--
>redhat-list mailing list
>unsubscribe mailto:redhat-list-request (AT) redhat (DOT) com?subject=unsubscribe
>
>>
>--
>redhat-list mailing list
>unsubscribe mailto:redhat-list-request (AT) redhat (DOT) com?subject=unsubscribe
>
>>

1/26/07, Lord of Gore <lordofgore (AT) logsoftgrup (DOT) rowrote:
katsumi liquer wrote:
and the installer is very intelligent. Their are cases where I think
CS/RHEL is better, but I have been let down by the redhat
architecture a
few times lately.

No rock at all but tell us about letting down, I am interested when it
failed you, maybe we'll avoid such problems.

Agreed. I'd be interested in hearing about your let downs as well. I
have a number of RHEL server at $work that I've been managing without
any issues as well a number of CS servers at customer sites that
haven't experienced any issues.

issue I have had with RHEL 4.x is closely related with VMware
Server -- RHEL does not seem to make a great platform for this.
Despite a lot of help from RH tech support, I have problems with
memory and swap, and in particular the 'M' out-of-memory killer.
Basically, in our case we have a RHEL 4.4 box with 14 GiB of main
memory and 10 GiB of swap; inside VMware Server is configured to have
12 GiB, configured with the potential to use swap when necessary. For
some reason, even when I set the kernel parameter to change/disable
the M killer behavior, it will still pop up at random times select a
process to terminate which is using the most memory. Generally, swap
is always %0 utilized when this happens. I have to be fair and say
that I *hate* the M killer, and that one single feature put a lot of
distrust between me and the RHEL kernel. Even after forcing the
feature off, it will still be triggered sometimes even when their is
still free memory and plenty of swap.

A second feature which I don't like about RHEL is that the syslog
daemon is permanently configured for event suppression -- meaning that
if a certain event repeats a certain number of times, syslog will
print out a message like: 'message repeated' -- and you can't disable
this behavior. it is all fine and good, except when you are trying to
get very accurate statistics from your syslog daemon, say for an IDS.
I talked to RH tech support about this, and they said that suppression
is there to protect your log file size, and that you can't disable it.

I know it is two small details, but in both cases they have been a
hindrance to the reason we chose RedHat as a platform in the first
place (stability, flexibility)

than that, RHEL has generally been very reliable, and I prefer
their network configuration structure to Ubuntu.

katsu

1/26/07, Gaddis, Jeremy L. <jeremy (AT) linuxwiz (DOT) netwrote:
1/26/07, Lord of Gore <lordofgore (AT) logsoftgrup (DOT) rowrote:
katsumi liquer wrote:
and the installer is very intelligent. Their are cases where I think
CS/RHEL is better, but I have been let down by the redhat
architecture a
few times lately.

No rock at all but tell us about letting down, I am interested when it
failed you, maybe we'll avoid such problems.

Agreed. I'd be interested in hearing about your let downs as well. I
have a number of RHEL server at $work that I've been managing without
any issues as well a number of CS servers at customer sites that
haven't experienced any issues.

katsumi liquer wrote:
issue I have had with RHEL 4.x is closely related with VMware
<snipped>

A second feature which I don't like about RHEL is that the syslog
daemon is permanently configured for event suppression -- meaning that
if a certain event repeats a certain number of times, syslog will
print out a message like: 'message repeated' -- and you can't disable
this behavior. it is all fine and good, except when you are trying to
get very accurate statistics from your syslog daemon, say for an IDS.
I talked to RH tech support about this, and they said that suppression
is there to protect your log file size, and that you can't disable it.

I recall that there is a kernel printk imposed the rate limit - I did
not think klogd or syslogd imposed any rate limit. It is configurable,
check with sysctl command :

# sysctl -a |grep printk_ratelimit
kernel.printk_ratelimit_burst = 10
kernel.printk_ratelimit = 5

From the file
/ in the
kernel-doc rpm:

printk_ratelimit:

Some warning messages are rate limited. printk_ratelimit specifies
the minimum length of time between these messages (in jiffies), by
default we allow one every 5 seconds.

A value of 0 will disable rate limiting.

printk_ratelimit_burst:

While long term we enforce one message per printk_ratelimit
seconds, we do allow a burst of messages to pass through.
printk_ratelimit_burst specifies the number of messages we can
send before ratelimiting kicks in.

Cheers
Michael

Hi Mike,

Wow -- that is cool!! I have not heard of that parameter before. I
think I could probably use that -- if I set the rate limit high
enough, perhaps it would prevent syslog from suppressing duplicate
messages. The problem though is that I am guessing this only applies
to messages coming from the kernel; other processes which talk to
syslog directly will probably still trigger suppression I'm guessing.
For example, several repeated failure messages from sshd will generate
this message:

"last message repeated"

This is basically the situation:

I have never been able to find the official RedHat solution to the
problem, short of rebuilding a custom syslog

katsu

1/30/07, Mike Kearey <mkearey (AT) redhat (DOT) comwrote:
katsumi liquer wrote:
issue I have had with RHEL 4.x is closely related with VMware
<snipped>

A second feature which I don't like about RHEL is that the syslog
daemon is permanently configured for event suppression -- meaning that
if a certain event repeats a certain number of times, syslog will
print out a message like: 'message repeated' -- and you can't disable
this behavior. it is all fine and good, except when you are trying to
get very accurate statistics from your syslog daemon, say for an IDS.
I talked to RH tech support about this, and they said that suppression
is there to protect your log file size, and that you can't disable it.
--
I recall that there is a kernel printk imposed the rate limit - I did
not think klogd or syslogd imposed any rate limit. It is configurable,
check with sysctl command :

# sysctl -a |grep printk_ratelimit
kernel.printk_ratelimit_burst = 10
kernel.printk_ratelimit = 5
>
>
>
>
From the file
/ in the
kernel-doc rpm:

printk_ratelimit:

Some warning messages are rate limited. printk_ratelimit specifies
the minimum length of time between these messages (in jiffies), by
default we allow one every 5 seconds.

A value of 0 will disable rate limiting.

printk_ratelimit_burst:

While long term we enforce one message per printk_ratelimit
seconds, we do allow a burst of messages to pass through.
printk_ratelimit_burst specifies the number of messages we can
send before ratelimiting kicks in.


>
>
>
Cheers
Michael

katsumi liquer wrote:
Hi Mike,

Wow -- that is cool!! I have not heard of that parameter before. I
think I could probably use that -- if I set the rate limit high
enough, perhaps it would prevent syslog from suppressing duplicate
messages. The problem though is that I am guessing this only applies
to messages coming from the kernel; other processes which talk to
syslog directly will probably still trigger suppression I'm guessing.
For example, several repeated failure messages from sshd will generate
this message:

"last message repeated"

This is basically the situation:

I have never been able to find the official RedHat solution to the
problem, short of rebuilding a custom syslog

Hi Again, sorry I was not very clear or helpful in my message ( I was in
a hurry :) )
- the kernel rate_limit is just for kernel messages
- syslog does suppress repeated messages. I don't know what the method
is for detecting the repeated messages The important point to be aware
of is that this is a 'syslog for Linux' feature, and not a Red Hat
specific thing. Also it's probably not considered a problem, that is why
there is not official Red Hat solution.

I'd consider syslog-ng as it's much more configurable. You won't have to
modify syslog at all, and pre-built packages are available for RHEL4.

Cheers
Michael