Hey everyone, I know, I'm an early bird, but Debian Jessie is right around the corner, and I just did an upgrade from 7 to 8 on a test server befor installing ISPConfig. I know there will be an official tutorial on the site, but does anyone already know some painpoints that we need to keep in mind? In another thread, it was already mentioned that suphp was removed (for good?), and I also encountered php-ming was removed as well. Anything else we have to keep in mind? Please feel free to use this thread as a collection of "best practices". Regards Zero EDIT: I got many messages regarding SNMP OIDs. Installing snmp-mibs-downloader fixed that.
ISPConfig works fine on debian testing, so there are no issues with jessie. The php-ming package can be left out.
@till, well, I just encountered that snmp-Problem which I only could solve through installing that additional package. Did you do a clean install of testing or an upgrade from Wheezy? Because maybe that is what's causing my problems. I would like to ask you to verify that as soon as you have time for that.
php5-ming, php5-ps, libapache2-ruby are not longer present in Jessie. Also SuPHP. You must remove this. for ruby you must install libapache2-passanger for dovecot you need also dovecot-lmtpd with this, i got it to work Take a look into: https://github.com/servisys/ispconfig_setup/tree/development/distros/debian8 I never hab problems with snmp, only with lmtpd.
Hi all One of my server upgraded itself to debian 8 and now apache lost all the virtualhosts anyone can help me?
What did you exactly. A dist-upgrade normally don't remove vhosts? Did you run ispconfig_update.sh after dist-upgrade and resync all?
webmin auto upgraded my server, now it's debian 8, apache appear to be broken. After manual upgrade of ispconfig I have this error Job for apache2.service failed. See 'systemctl status apache2.service' and 'journalctl -xn' for details. apache2.service loaded failed failed LSB: Apache2 web server
root@cloud001:~# systemctl status apache2.service ● apache2.service - LSB: Apache2 web server Loaded: loaded (/etc/init.d/apache2) Active: failed (Result: exit-code) since Sun 2015-04-26 15:56:13 CEST; 56s ago Process: 11902 ExecStop=/etc/init.d/apache2 stop (code=exited, status=0/SUCCESS) Process: 13490 ExecStart=/etc/init.d/apache2 start (code=exited, status=1/FAILURE) Apr 26 15:56:13 cloud001 apache2[13490]: The apache2 configtes... Apr 26 15:56:13 cloud001 apache2[13490]: Output of config test... Apr 26 15:56:13 cloud001 apache2[13490]: AH00548: NameVirtualH... Apr 26 15:56:13 cloud001 apache2[13490]: AH00526: Syntax error... Apr 26 15:56:13 cloud001 apache2[13490]: Invalid command 'FCGI... Apr 26 15:56:13 cloud001 apache2[13490]: Action 'configtest' f... Apr 26 15:56:13 cloud001 apache2[13490]: The Apache error log ... Apr 26 15:56:13 cloud001 systemd[1]: apache2.service: control ... Apr 26 15:56:13 cloud001 systemd[1]: Failed to start LSB: Apac... Apr 26 15:56:13 cloud001 systemd[1]: Unit apache2.service ente... Hint: Some lines were ellipsized, use -l to show in full.
I upgraded to Jessie via the Squeeze to Wheezy guide and got some errors. My plan was to reinstall it anyways so I didn't really care about the result. Quotas are not working and it looks like Amavis/Clamd and fail2ban is broken too. Everything is OK in the ISPConfig monitor and all websites works and are upgraded to PHP 5.6.7-1 using Apache 2.4.10. fail2ban[652]: Starting authentication failure monitor: fail2banERROR Found no accessible config files for 'filter.d/sasl' under /etc/fail2ban fail2ban[652]: ERROR Unable to read the filter fail2ban[652]: ERROR Errors in jail 'sasl'. Skipping... fail2ban[652]: failed! clamd[3363]: WARNING: lstat() failed on: /var/lib/amavis/tmp/amavis-20150426T035332-03443-v4Nxx3rQ/parts amavis[3443]: (03443-03) (!)run_av (ClamAV-clamd) FAILED - unexpected , output="/var/lib/amavis/tmp/amavis-20150426T035332-03443-v4Nxx3rQ/parts: lstat() failed: Permission denied. ERROR\n" amavis[3443]: (03443-03) (!)ClamAV-clamd av-scanner FAILED: CODE(0xb52eed8) unexpected , output="/var/lib/amavis/tmp/amavis-20150426T035332-03443-v4Nxx3rQ/parts: lstat() failed: Permission denied. ERROR\n" at (eval 99) line 905. amavis[3443]: (03443-03) (!)WARN: all primary virus scanners failed, considering backups quotaon.sh[238]: Checking quotas... systemd[1]: quota.service: main process exited, code=exited, status=1/FAILURE systemd[1]: Failed to start Check And Enable File System Quotas. systemd[1]: Unit quota.service entered failed state.
What would be the best practice for to migrate to debian8? I'm stable and happy with debian7 (7.8) but now that debian8 is stable I was thinking to migrate to it. I would really love to dist-upgrade rather than reinstall the whole system as I'm running some webistes for a couple of friends. What would you suggest to do? Which is the most effective way to backup ispconfig?
@bch: The config for jessie is different then the one for wheezy, so just doing the dist upgrade is not enough. We will release a perfect server guide for jessie today that contains all config steps for the new distribution. Never ugrade to a fresh released linux distribution immedaitely. One should always wait a month or two untill the new distribution has been stabilized and the last issues have been ironed out.
@till yes, that's probably the best advice (no irony here!) And an even better advice would be: Don't upgrade at all. I mean, the oldstable will be provided with updates for some more months. In the meantime, you could easily setup a NEW Server with a fresh installation and do a migration (a so-called three point migration would be best: 1. Create sites and mails on the new server 2. Move database to new server and reconfigure website configs to the new databases on the new server. 3. Change DNS-Settings and do a constant synching of website files) I encountered something strange. It seems that amavisd-new seems to stop listening on port 10024/tcp (IPv4) after some time, but still listens to the port on IPv6....either this is strange, or I haven't found the correct switch yet. Anyway, I will wait for the tutorial and make some tweaks to the config files where necessary. @bch: The clamav problem seems to be easily fixed: @ /etc/clamav/clamd.conf, change "AllowSupplementaryGroups" from false to true. Also, check if amavis is in the group clamav and vice versa.
@till: That's great. I just thought it would work out of the box because "testing" is supported, but of course that doesn't mean that it can be dist-upgraded without issues (and I did reinstall ISPConfig and reconfigure services). There is also valuable knowledge in the official upgrade documentation https://www.debian.org/releases/stable/amd64/release-notes/ch-upgrading.en.html (& chapter 5) - I guess you are aware of it.
I did the upgrade today from a fully up-to-date wheezy with dotdeb php 5.6 upgraded in wheezy (amd64) on a vps, and had the following problems: 1. initscripts wouldn't configure - I had to delete the /dev/shm symlink sand reconfigure initscripts which did finally configure and allowed ifupdown to configure and the rest of the updates to take place. spamassassin has a new conf file which required me to edit /etc/default/spamassassin and set Enable=1 - (the default is now 0) - to get is started again. Fail2ban and clamav-daemon still have problems, which may have to do with apache 2.4's changes from 2.2. A fatal error : no SASL authentication mechanisms installing cyrus-sasl2-dbg did NOT solve this error The ispconfig3 port 8080 site needs to have its vhost data modified so that it can be reached, as do all other sites hosted on the server or vps. The main difference in jessie for most users is the apache 2.2 to 2.4.x upgrade which requires you to change every sites'vhost declarations. Make sure you change any entries that refer to "testing" in your /etc/apt/sources.list to "stable". Here's a decent guide to upgrading from apache 2.2 to the 2.4 series (which is more like Nginx than 2.2 and can serve many more pages/hits than 2.2): https://www.linode.com/docs/securit...l-host-settings-from-apache-2-2-to-apache-2-4
See Jessie perfect server guide which has been released today for the nescessary changes. You have to run a ispconfig update after you upgraded to jessie so ispconfig can update all config files to the syntax of the newly installed software versions, there is no need to manually edit the ispconfig vhost file. On a ispconfig server, you just login to ispconfig, go to tools > resync and choose to resync all websites.
FYI: After update to jessie i cannot get quota running, when system starting, getting message "Failed to start Check and Enable File System Quotas See 'systemctl status quota.service' Which results to : quota.service - Check And Enable File System Quotas Loaded: loaded (/lib/systemd/system/quota.service; enabled) Active: failed (Result: exit-code) since Tue 2015-04-28 21:20:35 CEST; 1min 32s ago Docs: man:quotaon(8) Process: 214 ExecStart=/usr/share/quota/quotaon.sh (code=exited, status=1/FAILURE) Main PID: 214 (code=exited, status=1/FAILURE) Apr 28 21:20:35 vps000 quotaon.sh[214]: Checking quotas... Apr 28 21:20:35 vps000 systemd[1]: quota.service: main process exited, code=exited, status=1/FAILURE Apr 28 21:20:35 vps000 systemd[1]: Failed to start Check And Enable File System Quotas. Apr 28 21:20:35 vps000 systemd[1]: Unit quota.service entered failed state. I have tried reinstall quota , delete quota files, reinit quota, nothhing helps.... There is no hurry, i'm just testing if there will be possibility to update production servers with wheezy Also ISPConfig is not starting, but maybe after ISPC reinstall it will start work There was also some error messages about jailkit and sasl. clamav-daemon problemis solved with purge and reinstall package SNMP OIDs.problem which was filling apache error log solved with Installing snmp-mibs-downloader fas mentioned above
ISPConfig is not a dameon, so there is nothing to be started for ispconfig. In any case, yu have to run a ispconfig updater after a dist upgrade and choose to reconfigure services.
