Just a quick followup: After I added a second 512MB swap file on January 12th I haven't had an invoked-oom failure. I have had some high swap activity and will consider deleting the 512MB swap file and creating a 1024MB swap file. A search of an older /var/log/messages file shows that several processes were being killed by the oom manager, not just mysql. As performance isn't an issue I will change some of the mysql settings to reduce its memory footprint.

For a low volume site this basic linode server looks to be adequate to run my ClearOS 7 Home Essentials with Zarafa. If I needed more performance in a real work environment I would opt for the $20.00/month 2 core, 4GB RAM option.

Thanks Tony,

I have created another 512M swap file and set it at a lower priority. I will track usage and check for invoked-oom error messages. Here is a new free -m and cat /proc/swaps.

[root@nss init.d]# free -m
total used free shared buff/cache available
Mem: 1997 1693 33 45 270 200
Swap: 1023 446 577

[root@nss init.d]# cat /proc/swaps
Filename Type Size Used Priority
/dev/sdb partition 524284 457692 -1
/swapfile file 524284 0 -2

For Zarafa gurus out there, I would like to disable starting zarafa-ical, monitor and search if someone knows where they are started. I will try removing the scripts from /etc/rc.d/init.d to see if that works. Turns out that stopping the webapp web server also disables activesync so that needs to stay up. I wil research ways to minimize mysql memory usage (if that's possible).

This only seems to happen when I add one of the email accounts to a new computer and it starts downloading thousands of emails and attachments. Shouldn't need to much of that going forward.

I will track this for the next few days and report if this resolved the invoked-oom errors.

Here is the free command on my server:

[root@nss ~]# free -m
total used free shared buff/cache available
Mem: 1997 1598 37 7 361 335
Swap: 511 48 463

Yes, I am getting occasional invoked-oom messages in my messages log for mysql that correspond to the failures. It looks like mysql gets restarted but I suspect that one or more zarafa servers isn't. I will check the next time this happens.

I have used Zarafa successfully in several large organizations (with more capable local servers) and have been happy with it. Particularly because it supports activesync for Outlook 2013 and 2016. Not my email client of choice but my wife depends on it for tax and quickbooks work.

I do have the apache web server installed but, I believe, zarafa uses a sandboxed copy, not the general use marketplace install. I can remove that server and will try to see what other marketplace products I can uninstall. The zarafa server was getting heavy use as I moved thousands of emails in multiple folders from local storage to zarafa. That activity might have caused the failure.

I would still like to find out how to stop the zarafa servers mentioned in the original post from starting on reboot. Another option might be to find a way to test if specific servers are running and either start or stop them using a cron job. Will take some research.

Running Zarafa Community in a 2GB Linode Server

I am running ClearOS 7.2 on a basic Linode cloud server. As configured it has 1 cpu core and 2 GB of RAM. It works just fine most of the time. I have had three instances where I couldn't log into the server from my email application (thunderbird, webapp & outlook 2016). Using webapp I see a message 'Unknown MAPI Error: MAPI_E_DISK_ERROR'. If I reboot the server the error goes away. I sent a support request to Linode and received a response within two hours! The tech indicated the hardware checked out fine. The tech also suggested that the server is over-using its allotment of RAM and is dipping into SWAP. The next time it fails I will look to see if any of the zarafa servers have failed.

In the meantime, I don't need all of the zarafa servers. To reduce RAM usage I can manually stop the webapp service, the ical service, the search indexing and the quota service. A chkconfig --list shows zarafa-ical, monitor and search to be all off for all runlevels. Obviously they are starting. Is there a .conf where I can stop them from starting? Also, where can I stop the web server for webapp from starting on a reboot?

The main purpose of this ClearOS installation is to handle email. I will try to remove other servers and services to reduce the RAM footprint. It is not heavily used (2 email accounts at this time) but it needs to be reliable. Any suggestions helpful.

This issue is solved! Thanks for your input Nick. And special thanks to Ben Chambers. He debugged the ClearOS Linode script with me. I have highlighted the change to the script below. So, to spin up ClearOS 7.x on Linode:

Create a Centos 7 server. Don't do any updates to the basic install.
Execute the ClearOS 7 StackScript with the modification noted below. The script is on the Linode StackScript site and the entire script is present in this discussion. Ben will update that script to add the second 'yum --enablerepo=* clean all' once he gets authorization to make the change.
I executed the lines in the script line by line just to be sure they worked. I did not do the default networking lines, skipping to and executing the two sed lines and the rest of the script.
You can then log in to the ClearOS 7 server on your assigned IP address:81 and complete the setup just like you would on a physical server install.

Got to say that for $10.00 a month this seems like a great way to run ClearOS. Particularly if you spend time in an RV tooling around the country and don't want to task your grandchildren with rebooting your server after a power failure.

Thanks again everyone....

*** Change required to current StackScript on Linode for ClearOS 7:

Hi Nick,
On Linode you install a native centos 7 and then run the script below to upgrade it ClearOS 7. As noted above, this time I did a yum update to the native centos 7 before starting to run the script line by line, not as a script. After doing the yum update to the native centos 7 server I was able to run each of the lines in the through the service webconfig stop. I tried to install the apps one at a time. app-accounts worked. When I executed yum again for app-configuration-backup it failed as noted above. Here is the script at Linode I am using:

Latest update: I rebuilt the centos 7 server. I then did a: yum update
Different error. As before, everything worked through the service webconfig stop. I was able to do the yum install for app-accounts successfully. When I tried to do the yum install for app-configuration-backup I had it fail with an error:

Error: Package: 32:bind-libs-9.9.4-37.el7.x86_64 (clearos-centos)
Requires: bind-license = 32:9.9.4-37.el7
Installed: 32:bind-license-9.9.4-38.el7_3.noarch (@updates)
bind-license = 32:9.9.4-38.el7_3
Available: 32:bind-license-9.9.4-29.el7_2.1.noarch (clearos-centos-updates)
bind-license = 32:9.9.4-29.el7_2.1
Available: 32:bind-license-9.9.4-29.el7_2.2.noarch (clearos)
bind-license = 32:9.9.4-29.el7_2.2
Available: 32:bind-license-9.9.4-29.el7_2.3.noarch (clearos-centos-updates)
bind-license = 32:9.9.4-29.el7_2.3
Available: 32:bind-license-9.9.4-29.el7_2.4.noarch (clearos-centos-updates)
bind-license = 32:9.9.4-29.el7_2.4
Available: 32:bind-license-9.9.4-37.el7.noarch (clearos-centos)
bind-license = 32:9.9.4-37.el7

It seems that doing the yum update after initially bringing up the centos server changed things. Does clearcenter need to update its repos for a different version of bind?

Here is the entire yum install command and output:

[root@li588-30 ~]# yum --enablerepo=clearos-centos,clearos-centos-updates -y install app-configuration-backup
Loaded plugins: fastestmirror
Loading mirror speeds from cached hostfile
* clearos: mirror2-frankfurt.clearos.com
* clearos-centos: download2.clearsdn.com
* clearos-centos-updates: download2.clearsdn.com
* clearos-contribs: mirror2-frankfurt.clearos.com
* clearos-infra: mirror2-frankfurt.clearos.com
* clearos-updates: mirror2-frankfurt.clearos.com
Resolving Dependencies
--> Running transaction check
---> Package app-configuration-backup.noarch 1:2.2.2-1.v7 will be installed
--> Processing Dependency: app-configuration-backup-core = 1:2.2.2-1.v7 for package: 1:app-configuration-backup-2.2.2-1.v7.noarch
--> Processing Dependency: app-registration for package: 1:app-configuration-backup-2.2.2-1.v7.noarch
--> Running transaction check
---> Package app-configuration-backup-core.noarch 1:2.2.2-1.v7 will be installed
--> Processing Dependency: app-network-core >= 1:2.2.4 for package: 1:app-configuration-backup-core-2.2.2-1.v7.noarch
--> Processing Dependency: app-tasks-core for package: 1:app-configuration-backup-core-2.2.2-1.v7.noarch
---> Package app-registration.noarch 1:2.1.18-1.v7 will be installed
--> Processing Dependency: app-registration-core = 1:2.1.18-1.v7 for package: 1:app-registration-2.1.18-1.v7.noarch
--> Running transaction check
---> Package app-network-core.noarch 1:2.2.6-1.v7 will be installed
--> Processing Dependency: rp-pppoe >= 3.10-8.1 for package: 1:app-network-core-2.2.6-1.v7.noarch
--> Processing Dependency: traceroute for package: 1:app-network-core-2.2.6-1.v7.noarch
--> Processing Dependency: telnet for package: 1:app-network-core-2.2.6-1.v7.noarch
--> Processing Dependency: tcpdump for package: 1:app-network-core-2.2.6-1.v7.noarch
--> Processing Dependency: syswatch for package: 1:app-network-core-2.2.6-1.v7.noarch
--> Processing Dependency: iw for package: 1:app-network-core-2.2.6-1.v7.noarch
--> Processing Dependency: csplugin-procwatch for package: 1:app-network-core-2.2.6-1.v7.noarch
--> Processing Dependency: bridge-utils for package: 1:app-network-core-2.2.6-1.v7.noarch
--> Processing Dependency: bind-utils for package: 1:app-network-core-2.2.6-1.v7.noarch
--> Processing Dependency: avahi for package: 1:app-network-core-2.2.6-1.v7.noarch
---> Package app-registration-core.noarch 1:2.1.18-1.v7 will be installed
--> Processing Dependency: app-clearcenter-core >= 1:1.4.8 for package: 1:app-registration-core-2.1.18-1.v7.noarch
---> Package app-tasks-core.noarch 1:2.1.6-1.v7 will be installed
--> Running transaction check
---> Package app-clearcenter-core.noarch 1:2.1.14-1.v7 will be installed
--> Processing Dependency: yum-marketplace-plugin for package: 1:app-clearcenter-core-2.1.14-1.v7.noarch
--> Processing Dependency: csplugin-audit for package: 1:app-clearcenter-core-2.1.14-1.v7.noarch
--> Processing Dependency: app-suva-core for package: 1:app-clearcenter-core-2.1.14-1.v7.noarch
--> Processing Dependency: app-language-core for package: 1:app-clearcenter-core-2.1.14-1.v7.noarch
---> Package avahi.x86_64 0:0.6.31-17.el7 will be installed
---> Package bind-utils.x86_64 32:9.9.4-37.el7 will be installed
--> Processing Dependency: bind-libs = 32:9.9.4-37.el7 for package: 32:bind-utils-9.9.4-37.el7.x86_64
--> Processing Dependency: liblwres.so.90()(64bit) for package: 32:bind-utils-9.9.4-37.el7.x86_64
--> Processing Dependency: libisccfg.so.90()(64bit) for package: 32:bind-utils-9.9.4-37.el7.x86_64
--> Processing Dependency: libisccc.so.90()(64bit) for package: 32:bind-utils-9.9.4-37.el7.x86_64
--> Processing Dependency: libisc.so.95()(64bit) for package: 32:bind-utils-9.9.4-37.el7.x86_64
--> Processing Dependency: libdns.so.100()(64bit) for package: 32:bind-utils-9.9.4-37.el7.x86_64
--> Processing Dependency: libbind9.so.90()(64bit) for package: 32:bind-utils-9.9.4-37.el7.x86_64
---> Package bridge-utils.x86_64 0:1.5-9.el7 will be installed
---> Package csplugin-procwatch.x86_64 0:1.0-6.v7 will be installed
---> Package iw.x86_64 0:4.3-1.el7 will be installed
--> Processing Dependency: libnl.so.1()(64bit) for package: iw-4.3-1.el7.x86_64
---> Package rp-pppoe.x86_64 0:3.11-5.v7 will be installed
---> Package syswatch.noarch 0:7.1.3-1.v7 will be installed
---> Package tcpdump.x86_64 14:4.5.1-3.el7 will be installed
---> Package telnet.x86_64 1:0.17-60.el7 will be installed
---> Package traceroute.x86_64 3:2.0.22-2.el7 will be installed
--> Running transaction check
---> Package app-language-core.noarch 1:2.1.10-1.v7 will be installed
---> Package app-suva-core.noarch 1:2.2.1-1.v7 will be installed
--> Processing Dependency: suva-client for package: 1:app-suva-core-2.2.1-1.v7.noarch
---> Package bind-libs.x86_64 32:9.9.4-37.el7 will be installed
--> Processing Dependency: bind-license = 32:9.9.4-37.el7 for package: 32:bind-libs-9.9.4-37.el7.x86_64
---> Package csplugin-audit.x86_64 0:1.1-8.v7 will be installed
---> Package libnl.x86_64 0:1.1.4-3.el7 will be installed
---> Package yum-marketplace-plugin.noarch 0:2.1-12.v7 will be installed
--> Running transaction check
---> Package bind-libs.x86_64 32:9.9.4-37.el7 will be installed
--> Processing Dependency: bind-license = 32:9.9.4-37.el7 for package: 32:bind-libs-9.9.4-37.el7.x86_64
---> Package suva-client.x86_64 0:3.1-17.v7 will be installed
--> Finished Dependency Resolution
Error: Package: 32:bind-libs-9.9.4-37.el7.x86_64 (clearos-centos)
Requires: bind-license = 32:9.9.4-37.el7
Installed: 32:bind-license-9.9.4-38.el7_3.noarch (@updates)
bind-license = 32:9.9.4-38.el7_3
Available: 32:bind-license-9.9.4-29.el7_2.1.noarch (clearos-centos-updates)
bind-license = 32:9.9.4-29.el7_2.1
Available: 32:bind-license-9.9.4-29.el7_2.2.noarch (clearos)
bind-license = 32:9.9.4-29.el7_2.2
Available: 32:bind-license-9.9.4-29.el7_2.3.noarch (clearos-centos-updates)
bind-license = 32:9.9.4-29.el7_2.3
Available: 32:bind-license-9.9.4-29.el7_2.4.noarch (clearos-centos-updates)
bind-license = 32:9.9.4-29.el7_2.4
Available: 32:bind-license-9.9.4-37.el7.noarch (clearos-centos)
bind-license = 32:9.9.4-37.el7
You could try using --skip-broken to work around the problem
You could try running: rpm -Va --nofiles --nodigest
[root@li588-30 ~]#

I tried changing the nameservers to google with no change after rebuilding back to vanilla centos 7. Again, I ran the statements in the clearos7 script line by line using putty. Everything worked through the 'service webconfig stop' command.When I executed the next line to install only app-accounts, that worked. I then tried to install app-configuration-backup by itself. It failed with the 'Requested Range not Satisfied' error on several modules:

Error downloading packages:
3:traceroute-2.0.22-2.el7.x86_64: [Errno 256] No more mirrors to try.
avahi-libs-0.6.31-17.el7.x86_64: [Errno 256] No more mirrors to try.
GeoIP-1.5.0-11.el7.x86_64: [Errno 256] No more mirrors to try.
1:telnet-0.17-60.el7.x86_64: [Errno 256] No more mirrors to try.
32:bind-utils-9.9.4-38.el7_3.x86_64: [Errno 256] No more mirrors to try.
avahi-0.6.31-17.el7.x86_64: [Errno 256] No more mirrors to try.
32:bind-libs-lite-9.9.4-38.el7_3.x86_64: [Errno 256] No more mirrors to try.
32:bind-libs-9.9.4-38.el7_3.x86_64: [Errno 256] No more mirrors to try.
avahi-autoipd-0.6.31-17.el7.x86_64: [Errno 256] No more mirrors to try.
32:bind-license-9.9.4-38.el7_3.noarch: [Errno 256] No more mirrors to try.
iw-4.3-1.el7.x86_64: [Errno 256] No more mirrors to try.

Each of them tried download1 through 4.clearsdn.com.

I then tried to install only traceroute using: yum --enablerepo=clearos-centos,clearos-centos-updates -y install traceroute

yum tried all four mirrors and failed with the 'Range Not Satisfied' error.

yum -y install traceroute installed traceroute-2.0.19-5.el7.x86_64 instead of traceroute-2.0.22-2.el7.x86_64.

Are some of the dependencies not present in the repos or are the dependent versions not correct.

Not much good info on this error on Google.

ClearOS 7 Install on Linode fails

Trying to install ClearOS 7 Home Edition on a Linode server using: https://www.clearos.com/resources/documentation/clearos/content:en_us:7_linode instructions. After installing it via the StackScript for ClearOS 7 I am able to log in at https://1.2.3.4:81. When I click on Next I get a 404 error (Page Not Found). I have tried running the first line of the script:
rpm -Uvh http://download2.clearsdn.com/marketplace/cloud/7/noarch/clcearos-release-7-current.noarch.rpm
and get the message:
clearos-release >= 7-1.19 is needed by (installed) app-base-core-1:2.2.15-1.v7.noarch

Does this script need to be updated? The build process seemed to go too fast to have run the script needed to update centos 7 to ClearOS 7.