Topic: Version 3 of unofficial update script

[smeghead]

I changed the script to point to contribs cos when I last used it (a coupla dayas ago) there was a version mismatch between the script from swerts-knudsen and the files on contribs.

I figured it would be more relaible (tho poss less up to date) to get the script and files from the same place; the scripts point to contribs as the primary d/l location.

HTH

[gregswallow]

I tried the PHP admin thing, but I get "Document contains no data" message.
...
Please help  

Mophilly was on the right track - but watch your capitalization - https://www.example.com/phpmyadmin - that should work - /phpMyAdmin doesn't.

[soprom]

When I applied the script, I lost the Cyrus-imap addon. So, following gregswallow's advice, I excluded updating apache (because of db4 - cyrus is depending on db3)

I've downloaded everything and applied the script manually, checking for cyrus to work correctly between each step. Since I did not reboot after each step, the test might not be good.

I skipped apache because of db4 and I also skipped
securemail because e-smith-imap is needed by e-smith-ssl-imap.

It resulted with a no go! Trying to login:
cyradm -user cyrus localhost
IMAP Password:  <<-- same pw as before upgrade
Login failed: authentication failure at /usr/lib/perl5/site_perl/5.6.1/i386-linux/Cyrus/IMAP/Admin.pm line 114
cyradm: cannot authenticate to server with  as cyrus

Maybe the double-bounce kit is the problem...

For more information, here's an overview of the installation of the cyrus-imap server:

rpm -e e-smith-imap
rpm -e devecot
/sbin/e-smith/config setprop imap status disabled
/sbin/e-smith/config setprop popd status disabled
cyrus-imapd-2.1.12-9.i386.rpm
cyrus-imapd-devel-2.1.12-9.i386.rpm
cyrus-imapd-utils-2.1.12-9.i386.rpm
cyrus-sasl-2.1.12-2.i386.rpm
cyrus-sasl-devel-2.1.12-2.i386.rpm
cyrus-sasl-gssapi-2.1.12-2.i386.rpm
cyrus-sasl-md5-2.1.12-2.i386.rpm
cyrus-sasl-ntlm-2.1.12-2.i386.rpm
cyrus-sasl-plain-2.1.12-2.i386.rpm
db3x-3.2.9-4.i386.rpm
perl-Cyrus-2.1.12-9.i386.rpm

[girkers]

Thanks Greg, but alas poor Yorick it does not work.  I think there may be an issue with my SSL not being turned on, but I do not where to turn.

  for going off topic, sorry.

[gregswallow]

Maybe the double-bounce kit is the problem...

I think you are on the right track.  You'll have to do the updates one by one and see what is causing the problem.  I think only the mysql update requires a reboot, the rest you should be able to see the effect right away.

[soprom]

I tried the script without the double-bounce part and the cyrus-imap is still broken. Too bad. I guess the package for cyrus wont survive the coming versions of the SME. It's a good thing this script was available. I can know focus on an other distro if I want a cyrus server. The Kolab2 seems to be most interesting.

[gregswallow]

I changed the script to integrate the mysql and php updates - a couple links in those scripts were dead.  It is better this way so now we have error checking for the downloads of those rpms.  Also updated the links to mysql 4.0.22 rpms.

[Henk]

After updating apache to 1.3.33 on SME 6.01.01 using the rpm's from the update script, my webserver suddenly went down after a couple of days.

Some more people (from a dutch site, minddigger.com) also updated apache to 1.3.33 and have troubles with it. The apache 1.3.33 rpm's from norlug seem to be unstable (at least on SME). Some people cannot access the servermanager anymore and have to restart https-admin before it works again.

Anyone else having problems with the apache update?

[gregswallow]

I did notice having to run:

Code: [Select]

service httpd-admin restart
...to access the server manager.

I just have that update installed on a test server, but really havent tested it too much.  The 1.3.31 or 32 version from norlug I have on another server though, and is working fine.

Maybe we should switch to using the apache rpms from fedoralegacy - eg.:
http://download.fedoralegacy.org/redhat/7.3/updates/i386/apache-1.3.27-6.legacy.i386.rpm

That one is dated Nov. 17th.   I don't know if it is as up to date as 1.3.33, the only notes about it I found are: http://fedoralegacy.org/updates/RH7.3/2004-12-03_FLSA_2004_2148__Updated_httpd__apache_and_mod_ssl_packages_fix_security_issues.html

[koopmann]

I upgradede with the script on a fresh SME601. After that I made a restore created on my former SME600 to restore users etc. Now I daily get the error from cron:
Cron <root@serv> /etc/clamscan
LibClamAV Warning: Unsupported multipart format knowbot' - report to bugs@clamav.net
LibClamAV Warning: Unsupported multipart format knowbot' - report to bugs@clamav.net
LibClamAV Warning: Unsupported multipart format knowbot' - report to bugs@clamav.net
LibClamAV Warning: Multipart MIME message contains no boundaries
LibClamAV Warning: messageFindArgument: no '=' sign found in MIME header
LibClamAV Warning: Multipart MIME message contains no boundaries
LibClamAV Warning: Unknown encoding type "plain" - report to bugs@clamav.net
LibClamAV Warning: Unknown encoding type "plain" - report to bugs@clamav.net
LibClamAV Warning: Ignoring empty field in " charset="
LibClamAV Warning: Ignoring empty field in " charset="
LibClamAV Warning: Unsupported multipart format knowbot' - report to bugs@clamav.net
LibClamAV Warning: Unsupported multipart format knowbot' - report to bugs@clamav.net.

Has anybody any idea to fix this.

Birger Koopmann

[gregswallow]

I upgradede with the script on a fresh SME601. After that I made a restore created on my former SME600 to restore users etc. Now I daily get the error from cron:
Cron <root@serv> /etc/clamscan
LibClamAV Warning: Unsupported multipart format knowbot' - report to bugs@clamav.net

Regarding the error:
http://www.google.ca/search?hl=en&q=Unsupported+multipart+format+%60knowbot%27+&btnG=Google+Search&meta=
...(try the 2nd url...it seems to be a bug in clamav .80 that is fixed in the upcoming .81 already.

The CLAMAV script is from:
http://sme.swerts-knudsen.com/howtos/howto_22.htm#Revision_History
...keep an eye on his page for an upgrade - you can run the script (antivirus_install.sh) again when he has .81 and that should fix your error.

[Lum-chan]

Hmm after installing this my FetchMail won't distribute mail to the local users anymore. Worked fine before.
What went wrong?

Is there something I overlooked?

[gregswallow]

Is there something I overlooked?

Yes - the warnings:

Code: [Select]

   echo '| This script is *no* official update and intended only for fresh     |'
    echo '| installations of SME 6.0.1-01 but *may* work with modified          |'
    echo '| SME 6.01 installations, too.                                        |'
    echo '|                                                                     |'
    echo '| No express or implied warranties are provided and its usage is at   |'
    echo '| your own risk.    

But seriously, take a look at http://www.schirrms.net/sme/SMEFetchMail.php for information on the Fetchmail contributions that this script installs.  That would likely be what broke your existing setup.  Check the server-manager if all your old settings are still there.  Maybe provide some more information if you would like some more help.

[Lum-chan]

Yes - the warnings:

Code: [Select]

   echo '| This script is *no* official update and intended only for fresh     |'
    echo '| installations of SME 6.0.1-01 but *may* work with modified          |'
    echo '| SME 6.01 installations, too.                                        |'
    echo '|                                                                     |'
    echo '| No express or implied warranties are provided and its usage is at   |'
    echo '| your own risk.    

But seriously, take a look at http://www.schirrms.net/sme/SMEFetchMail.php for information on the Fetchmail contributions that this script installs.  That would likely be what broke your existing setup.  Check the server-manager if all your old settings are still there.  Maybe provide some more information if you would like some more help.

Ah oke, But that was quite the contrib which was installed and running before I applied the smeplus.sh script.
Maybe that's what went wrong. In the mean time I have re-installed a fresh copy of SME and check the scripts for contribs I want and install them manually.

Thanks for the advise!

I learned at least one thing or actually two: Always make a backup and learn to read the warnings

BTW: Oh well... I'm just starting with SME and have to get used to SME... Used Freesco before so this is quite new to me.

[Denbert]

Hi there,

I’ve used the smeplus.sh script to update my server, at I love all the new contribs in the server-manager.

After the reboot my server still says SME Server 6.0-04 in the server-manager

I this correct, or did I made any mistake?

There where NO errors written during the update, the script ran like a charm

Cheers