[BlueOnyx:24427] Re: strange issue vsite and hostname
Hello. I also experienced this problem the other day. The environment is 5209R. In my case, the result was a mismatch between the host name in the Let's Encrypt certificate and the reverted host name by reboot, causing the sendmail command to fail to send mail. # grep 'Data format error' /var/log/maillog Oct xx xx:xx:xx hostxx sendmail[18653]: 09RFKSvw0186xx: to=em...@address.xx, ctladdr= (1001/100), delay=00:00:00, xdelay=00:00:00, mailer=relay, pri=30210, relay=[127.0.0.1] [127.0.0.1], dsn=5.6.0, stat=Data format error ... When I checked the uptime, I found that there was a Data format error between the reboot and the time I noticed it. Thanks, 2020年10月28日(水) 5:05 Gerrit Haas : > > Hi all, > we are currently having trouble with one particular system. > The system hostname (TCP/IP Settings within server side) is -let's just say > abc.def.net. > From our documentation, the hostname *should* be abc.xyz.net. > On the server exists a vsite abc.def.net that is actually the same as the > hostname. > I did not expect this to actually work. We do not know, how it got there. > (Bug, Customer changed it,...) > > We tried changing the hostname back to abc.xyz.net. This works in the first > place, but after a reboot of the machine hostname abc.def.net is back. > Also it seems that the old name still resides in some configuration files in > /etc (admserv, hosts, ...) > > We are at this point a little clueless how to proceed and fix this. > This is messing up some certififaces or mailing stops working - which you > would expect. > > We also tried editing the server settings with ccceclient. (set 1."" > hostname="abc.xyz.net") - Same thing it will not survive any reboot. > Do you have any hints or ideas? > > Thank you all in advance! > Cheers > Gerrit > > > > Mit freundlichen Grüßen aus Bad Vilbel > > Gerrit Haas > Systemadministrator > > > blackpoint GmbH – Friedberger Straße 106b – 61118 Bad Vilbel > > Tel.: +49 6101 65788 32 > IT-Support: +49 6101 65788 - 30 > Fax: +49 6101 65788 - 99 > eMail: gerrit.h...@blackpoint.de > Tel. Rufbereitschaft (Außerhalb der Arbeitszeiten) +49 6101 65788 - 40 > > Vertretungsberechtigt Dirk Estenfeld und Mario Di Rienzo HRB 50093 Frankfurt > am Main USt.-IdNr. de210106871 > > Besuchen Sie uns im Internet unter www.blackpoint.de > Problemlos Domains registrieren: www.edns.de > Einfach und günstig Daten sichern: Veeam Cloud Connect > > > Confidentiality Notice: > This e-mail message, including any attachments,is for the sole use of the > intended recipient(s) and may contain confidential and privileged > information. Any unauthorized review, use, disclosure or distribution is > prohibited. If you are not the intended recipient, please contact the sender > by reply e-mail and destroy all copies of the original message. > ___ > Blueonyx mailing list > Blueonyx@mail.blueonyx.it > http://mail.blueonyx.it/mailman/listinfo/blueonyx ___ Blueonyx mailing list Blueonyx@mail.blueonyx.it http://mail.blueonyx.it/mailman/listinfo/blueonyx
[BlueOnyx:24426] 5210 Preview sites
The "Preview Site" feature doen't seem to work for me. I only get a 404. Any ideas where to look? ___ Blueonyx mailing list Blueonyx@mail.blueonyx.it http://mail.blueonyx.it/mailman/listinfo/blueonyx
[BlueOnyx:24424] strange issue vsite and hostname
Hi all, we are currently having trouble with one particular system. The system hostname (TCP/IP Settings within server side) is -let's just say abc.def.net. >From our documentation, the hostname *should* be abc.xyz.net. On the server exists a vsite abc.def.net that is actually the same as the hostname. I did not expect this to actually work. We do not know, how it got there. (Bug, Customer changed it,...) We tried changing the hostname back to abc.xyz.net. This works in the first place, but after a reboot of the machine hostname abc.def.net is back. Also it seems that the old name still resides in some configuration files in /etc (admserv, hosts, ...) We are at this point a little clueless how to proceed and fix this. This is messing up some certififaces or mailing stops working - which you would expect. We also tried editing the server settings with ccceclient. (set 1."" hostname="abc.xyz.net") - Same thing it will not survive any reboot. Do you have any hints or ideas? Thank you all in advance! Cheers Gerrit Mit freundlichen Grüßen aus Bad Vilbel Gerrit Haas Systemadministrator blackpoint GmbH Friedberger Straße 106b 61118 Bad Vilbel Tel.: +49 6101 65788 32 IT-Support: +49 6101 65788 - 30 Fax: +49 6101 65788 - 99 eMail: gerrit.h...@blackpoint.de Tel. Rufbereitschaft (Außerhalb der Arbeitszeiten) +49 6101 65788 - 40 Vertretungsberechtigt Dirk Estenfeld und Mario Di Rienzo HRB 50093 Frankfurt am Main USt.-IdNr. de210106871 Besuchen Sie uns im Internet unter www.blackpoint.de Problemlos Domains registrieren: www.edns.de Einfach und günstig Daten sichern: Veeam Cloud Connect Confidentiality Notice: This e-mail message, including any attachments,is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. smime.p7s Description: S/MIME cryptographic signature ___ Blueonyx mailing list Blueonyx@mail.blueonyx.it http://mail.blueonyx.it/mailman/listinfo/blueonyx
[BlueOnyx:24425] Re: strange issue vsite and hostname
Hi Gerrit,
> From our documentation, the hostname *should* be abc.xyz.net.
> On the server exists a vsite abc.def.net that is actually the same as the
> hostname.
> I did not expect this to actually work. We do not know, how it got there.
> (Bug, Customer changed it,...)
Yeah, this sure causes problems.
> We tried changing the hostname back to abc.xyz.net. This works in the first
> place, but after a reboot of the machine hostname abc.def.net is back.
> Also it seems that the old name still resides in some configuration files in
> /etc (admserv, hosts, ...)
On reboot the CCE constructor /usr/sausalito/constructor/base/system
uses the Perl module Sys::Hostname::FQDN to determine the servers host
and domain name this way:
($name,$aliases,$addrtype,$length,@addrs)=gethostinfo();
$myhost = short();
$fqdn = fqdn();
@hlist = split(/\s/, $aliases) ;
foreach $line (@hlist) {
if ($line =~ m/^$myhost\.(.*)$/ig ) {
unless (($line =~ m/localhost/ig) || ($line =~ m/localdomain/ig)) {
$fqdn = $line;
}
}
}
$mydomain = $fqdn;
$mydomain =~ s/^$myhost\.//;
if ( $myhost eq "" ) {
$myhost = "localhost";
}
if ( $mydomain eq "" ) {
$mydomain = "localdomain";
}
So we end up with either the real host and domain name, or
"localdomain.localhost".
To see what the system has stored in regards to the host and domain name
you can use these three commands:
uname -a
hostnamectl
nmcli g hostname
This info is then inserted into CODB and a handler runs the required
commands to tell the system its new name.
HOWEVER: CentOS 7 and CentOS 8 are a bit unique there. In the past the
"hostname" command was sufficient to set a new hostname. But with the
introduction of "static", "transient" and "pretty" hostnames (which can
all be somewhat different from each other) things got messy and just
using "hostname" to set the hostname doesn't stick all that well.
I just published YUM updates for 5209R and 5210R that now use *all* the
bloody methods with which one can set the hostname in the hope that if
we fling enough shit, some will eventually keep sticking to the wall.
Here are the commands that we're using for that:
/bin/hostname $fqdn
/usr/bin/hostnamectl set-hostname $fqdn
/usr/bin/hostnamectl set-hostname $fqdn --static
/usr/bin/hostnamectl set-hostname $fqdn --transient
/usr/bin/nmcli g hostname $fqdn
/usr/bin/systemctl restart systemd-hostnamed
But like said: Do a "yum clean all" and "yum update" and then once you
change the hostname via the GUI it should finally stick even through
reboots.
--
With best regards
Michael Stauber
___
Blueonyx mailing list
Blueonyx@mail.blueonyx.it
http://mail.blueonyx.it/mailman/listinfo/blueonyx
[BlueOnyx:24423] Re: 5210 Database management
> -Ursprüngliche Nachricht- > Von: Blueonyx Im Auftrag von > Meaulnes Legler @ MailList > Gesendet: Dienstag, 27. Oktober 2020 11:55 > An: blueonyx@mail.blueonyx.it > Betreff: [BlueOnyx:24419] Re: 5210 Database management > > On 27.10.20 10:36, Florian Arzberger wrote: > > a) > > Is there a way to find out which (if any) vsite a given db is tied to? > > if the name of the database isn't self-explanatory, I guess you have to dig > into the configuration files of the specific application running on a vsite... That's exactly what i'd like to avoid... > > b) > > Is there a way to tie an existing db to an existing vsite without > > deleting/re-creating? > > what are you trying to do? don't break anything!:-) > I have some databases that have been manually created (not using the BO GUI) and would like to get those tied to certain vsites. ___ Blueonyx mailing list Blueonyx@mail.blueonyx.it http://mail.blueonyx.it/mailman/listinfo/blueonyx
[BlueOnyx:24422] Re: 5210 Issue with cmuImport
I will once i am rid of that old BlueQuartz machine... ;) -Ursprüngliche Nachricht- Von: Blueonyx Im Auftrag von Michael Stauber Gesendet: Dienstag, 27. Oktober 2020 17:05 An: blueonyx@mail.blueonyx.it Betreff: [BlueOnyx:24421] Re: 5210 Issue with cmuImport Hi Florian, > When importing sites with cmuImport, the default index.html from > /etc/skel/... gets copied over upon vsite creation. This creates an > issue if the imported site does not habe an index.html but something > like index.php as starting page. Should be an easy fix ;). Do yourself a favor and ditch CMU. Use Easy-Migrate instead. Especially when migrating to 5210R. Here is a guide on how to use Easy-Migrate: https://www.blueonyx.it/easy-migrate It migrates everything. Including DNS and databases. -- With best regards Michael Stauber ___ Blueonyx mailing list Blueonyx@mail.blueonyx.it http://mail.blueonyx.it/mailman/listinfo/blueonyx ___ Blueonyx mailing list Blueonyx@mail.blueonyx.it http://mail.blueonyx.it/mailman/listinfo/blueonyx
[BlueOnyx:24421] Re: 5210 Issue with cmuImport
Hi Florian, > When importing sites with cmuImport, the default index.html from > /etc/skel/... gets copied over upon vsite creation. This creates an issue if > the imported site does not habe an index.html but something like index.php > as starting page. Should be an easy fix ;). Do yourself a favor and ditch CMU. Use Easy-Migrate instead. Especially when migrating to 5210R. Here is a guide on how to use Easy-Migrate: https://www.blueonyx.it/easy-migrate It migrates everything. Including DNS and databases. -- With best regards Michael Stauber ___ Blueonyx mailing list Blueonyx@mail.blueonyx.it http://mail.blueonyx.it/mailman/listinfo/blueonyx
[BlueOnyx:24420] AV-Spam Issues
Hi All, On 5210R. After the update the spam scores revert back to default [5.00] even though CCE shows a different value eg;10.00. Is there a way to reset all vsites and users spam score from the command line? Also see lots of these in maillogs since the update: Oct 27 10:45:07 ds1 sendmail[3136234]: NOQUEUE: SYSERR(smmsp): /etc/mail/sendmail.cf: line 167: fileclass: cannot open '/etc/mail/virthosts': Permission denied Also noticed that Server Management / Email / Advanced / [https://host.domain:81/email/emailsettings] Block Email From Hosts/Domains & Block Email From Users Won't save the first time you add a domain or email addy. Sometimes you need to add it 4-5 times before it saves. This did it before the update last nite. -- --- TIA David Hahn - PageKeeper Service ___ Blueonyx mailing list Blueonyx@mail.blueonyx.it http://mail.blueonyx.it/mailman/listinfo/blueonyx
[BlueOnyx:24414] Some nginx updates last night? nginx do not start with errors.
Hello,
at one server with nginx as front proxy nginx was not running after updates
this morning.
nginx: [emerg] "location" directive is not allowed here in
/etc/nginx/conf.d/default.conf:50
nginx: configuration file /etc/nginx/nginx.conf test failed
line 50/51/52
location ~ ^/libImage/*.*$ {
root /usr/sausalito/ui/web/;
}
I did a comment before lines 50, 51, 52 and 61.
Afterwards nginx did start.
I did find this feature at two different servers with activated nginx.
Please check and fix.
Thank you and best regards,
Dirk
blackpoint GmbH - Friedberger Straße 106b - 61118 Bad Vilbel
___
Blueonyx mailing list
Blueonyx@mail.blueonyx.it
http://mail.blueonyx.it/mailman/listinfo/blueonyx
[BlueOnyx:24419] Re: 5210 Database management
On 27.10.20 10:36, Florian Arzberger wrote: a) Is there a way to find out which (if any) vsite a given db is tied to? if the name of the database isn't self-explanatory, I guess you have to dig into the configuration files of the specific application running on a vsite... so for RoundCube, for instance, that runs a separate database for each vsite (e.g. nwa_0GnFGYyH4IWq), one can read its name in /roundcube/config/config.inc.php hope that answers your question b) Is there a way to tie an existing db to an existing vsite without deleting/re-creating? what are you trying to do? don't break anything!:-) best regards で⊃ Meaulnes Legler Zurich, Switzerland +41¦0 44 260-1660 ___ Blueonyx mailing list Blueonyx@mail.blueonyx.it http://mail.blueonyx.it/mailman/listinfo/blueonyx ___ Blueonyx mailing list Blueonyx@mail.blueonyx.it http://mail.blueonyx.it/mailman/listinfo/blueonyx
[BlueOnyx:24418] 5210 Database management
a) Is there a way to find out which (if any) vsite a given db is tied to? b) Is there a way to tie an existing db to an existing vsite without deleting/re-creating? ___ Blueonyx mailing list Blueonyx@mail.blueonyx.it http://mail.blueonyx.it/mailman/listinfo/blueonyx
[BlueOnyx:24417] 5210 Issue with cmuImport
When importing sites with cmuImport, the default index.html from /etc/skel/... gets copied over upon vsite creation. This creates an issue if the imported site does not habe an index.html but something like index.php as starting page. Should be an easy fix ;). ___ Blueonyx mailing list Blueonyx@mail.blueonyx.it http://mail.blueonyx.it/mailman/listinfo/blueonyx
[BlueOnyx:24416] Re: 5210 "-" char in DB-names not allowed anymore?
Your work is greatly appreciated. I was not shure if the mailing list is the right place to report issues like this. -Ursprüngliche Nachricht- Von: Blueonyx Im Auftrag von Michael Stauber Gesendet: Dienstag, 27. Oktober 2020 03:22 An: blueonyx@mail.blueonyx.it Betreff: [BlueOnyx:24411] Re: 5210 "-" char in DB-names not allowed anymore? Hi Florian, > When trying to manually create a database containing "-", the BO GUI > throws an error. Is that intended? No, this was an accident due to an incorrect regular expression in the Schema file. This has just been fixed in a YUM update for 5209R and 5210R. Many thanks for bringing this to my attention. And also for the recent donation. :-) -- With best regards Michael Stauber ___ Blueonyx mailing list Blueonyx@mail.blueonyx.it http://mail.blueonyx.it/mailman/listinfo/blueonyx ___ Blueonyx mailing list Blueonyx@mail.blueonyx.it http://mail.blueonyx.it/mailman/listinfo/blueonyx
[BlueOnyx:24415] Re: Some nginx updates last night? nginx do not start with errors.
Hi Dirk, Yeah, there was a larger set of updates tonight for 5209R and 5210R. Including an update for base-nginx. > at one server with nginx as front proxy nginx was not running after > updates this morning. > > nginx: [emerg] "location" directive is not allowed here in > /etc/nginx/conf.d/default.conf:50 > > nginx: configuration file /etc/nginx/nginx.conf test failed I just checked. Yes, the 5209R updates had that issue and I just published another set of updates to correct this. The 5210R updates were fine. The issue weren't line 50, 51 and 52, but the extra closing bracket before that block. When you do a "yum clean all" and "yum update" to get the latest fixes the default.conf will be corrected back to a working version with everything in place as it should be. Many thanks for the heads up. -- With best regards Michael Stauber ___ Blueonyx mailing list Blueonyx@mail.blueonyx.it http://mail.blueonyx.it/mailman/listinfo/blueonyx
