Clickhouse: Can't start clickhouse --help

[jgshzy]

os: Ubuntu 16.04.3 LTS n l
clickhouse version: 1.1.54342

I use " sudo service clickhouse-server start" . but not start.

so I use "systemctl status clickhouse-server.service" show message:

● clickhouse-server.service - ClickHouse Server (analytic DBMS for big data)
Loaded: loaded (/etc/systemd/system/clickhouse-server.service; enabled; vendor preset: enabled)
Active: activating (auto-restart) (Result: exit-code) since Thu 2018-03-01 22:31:55 CST; 21s ago
Process: 16057 ExecStart=/usr/bin/clickhouse-server --config=/etc/clickhouse-server/config.xml (code=exited, status=232/ADDRESS_FAMILIES)
Main PID: 16057 (code=exited, status=232/ADDRESS_FAMILIES)

Mar 01 22:31:55 ubuntu163 systemd[1]: clickhouse-server.service: Unit entered failed state.
Mar 01 22:31:55 ubuntu163 systemd[1]: clickhouse-server.service: Failed with result 'exit-code'.

[godjil]

Конфиг в студию!
Sho me Your config

[derekperkins]

I am seeing this too in testing. ClickHouse worked with all the defaults after installing with apt-get on Debian 9, but then I saw this error when I changed the location of the logs.

        <log>/mnt/resource/chdata/clickhouse-server.log</log>
        <errorlog>/mnt/resource/chdata/clickhouse-server.err.log</errorlog>

[alexey-milovidov]

ADDRESS_FAMILIES means that this is related to IPv4 vs IPv6 addresses in configuration.

Typical Linux system have support both for IPv4 and IPv6 and you can use :: wildcard address in config to listen.

But sometimes (in some cloud providers) they have a kernel with explicitly disabled IPv4 or IPv6.
In this case you have to edit the configuration (look at both config.xml and users.xml) to use only supported address families.

[alexey-milovidov]

I saw this error when I changed the location of the logs

Probably you have different error, could you please show it?

[derekperkins]

I figured it out, it was just a permissions error, but it also shows that error if the directory doesn't exist. I had to run chown clickhouse:clickhouse to fix it.

[PolinaIP]

I am seeing this too. What I do wrong?

lickhouse-server.service - ClickHouse Server (analytic DBMS for big data)
   Loaded: loaded (/etc/systemd/system/clickhouse-server.service; enabled; vendor preset: enabled)
   Active: activating (auto-restart) (Result: exit-code) since Thu 2018-08-09 13:35:27 UTC; 18s ago
  Process: 6945 ExecStart=/usr/bin/clickhouse-server --config=/etc/clickhouse-server/config.xml (code=exited, status=70)
  Process: 6941 ExecStartPre=/bin/chown clickhouse:clickhouse -R /etc/clickhouse-server (code=exited, status=0/SUCCESS)
  Process: 6938 ExecStartPre=/usr/bin/chown clickhouse:clickhouse -R /etc/clickhouse-server (code=exited, status=0/SUCCESS)
 Main PID: 6945 (code=exited, status=70)

[PolinaIP]

the problem is solved! I just had to change the user to root
in /etc/systemd/system/clickhouse-server.service

[healiseu]

I got the same problem with @jgshzy, but for a different reason. The first time I installed clickhouse the service started without any error. But then I decided to move clickhouse data repository to an SSD disk. I modified paths at
/etc/clickhouse-server/config.xml

<!-- Path to data directory, with trailing slash. -->
    <path>/dbstore/clickhouse/</path>

    <!-- Path to temporary data for processing hard queries. -->
    <tmp_path>/dbstore/clickhouse/tmp/</tmp_path>

    <!-- Directory with user provided files that are accessible by 'file' table function. -->
    <user_files_path>/dbstore/clickhouse/user_files/</user_files_path>

then I restarted the server and I got

sudo service clickhouse-server status
● clickhouse-server.service - ClickHouse Server (analytic DBMS for big data)
   Loaded: loaded (/etc/systemd/system/clickhouse-server.service; enabled; vendor preset: enabled)
   Active: activating (auto-restart) (Result: exit-code) since Thu 2018-09-27 21:04:14 EEST; 4s ago
  Process: 7774 ExecStart=/usr/bin/clickhouse-server --config=/etc/clickhouse-server/config.xml (code=exited, status=203/EXEC)
  Process: 7770 ExecStartPre=/bin/chown clickhouse:clickhouse -R /etc/clickhouse-server (code=exited, status=0/SUCCESS)
  Process: 7767 ExecStartPre=/usr/bin/chown clickhouse:clickhouse -R /etc/clickhouse-server (code=exited, status=203/EXEC)
 Main PID: 7774 (code=exited, status=203/EXEC)

Sep 27 21:04:14 Tornado systemd[1]: clickhouse-server.service: Unit entered failed state.
Sep 27 21:04:14 Tornado systemd[1]: clickhouse-server.service: Failed with result 'exit-code'

Then I tried several things with permissions, users, etc and eventually it was messed up completely so I decided to remove it and re-install it. I removed all clickhouse* directories first with mlocate and installed it again according to your quick start guide. But I noticed warnings and failure of something, see below

The following NEW packages will be installed
  clickhouse-client clickhouse-server
0 to upgrade, 2 to newly install, 0 to remove and 356 not to upgrade.
Need to get 0 B/18,4 kB of archives.
After this operation, 99,3 kB of additional disk space will be used.
Selecting previously unselected package clickhouse-client.
dpkg: warning: files list file for package 'clickhouse-common-static' missing; assuming package has no files currently installed
(Reading database ... 333742 files and directories currently installed.)
Preparing to unpack .../clickhouse-client_18.12.17_all.deb ...
Unpacking clickhouse-client (18.12.17) ...
Selecting previously unselected package clickhouse-server.
Preparing to unpack .../clickhouse-server_18.12.17_all.deb ...
Unpacking clickhouse-server (18.12.17) ...
Processing triggers for systemd (229-4ubuntu21.2) ...
Processing triggers for ureadahead (0.100.0-19) ...
ureadahead will be reprofiled on next reboot
Setting up clickhouse-client (18.12.17) ...
Setting up clickhouse-server (18.12.17) ...
ClickHouse init script has migrated to systemd. Please manually stop old server and restart the service: sudo killall clickhouse-server && sleep 5 && sudo service clickhouse-server restart
Synchronizing state of clickhouse-server.service with SysV init with /lib/systemd/systemd-sysv-install...
Executing /lib/systemd/systemd-sysv-install enable clickhouse-server
Failed to set capabilities on file `/usr/bin/clickhouse' (Invalid argument)
The value of the capability argument is not permitted for a file. Or the file is not a regular (non-symlink) file
Cannot set 'net_admin' capability for clickhouse binary. This is optional. Taskstats accounting will be disabled. To enable taskstats accounting you may add the required capability later manually.
Processing triggers for systemd (229-4ubuntu21.2) ...
Processing triggers for ureadahead (0.100.0-19) ...

Then I did
sudo service clickhouse-server start
and unfortunately I got the same error, can you help please ?

I also noticed that all binaries /usr/bin/clickhouse-* point to clickhouse and I cannot find clickhouse. It looks like the clickhouse package is not installed correctly

PS: It will also help others if you write a guide with instructions on how to move safely data to another filesystem e.g. mounted on an SSD.

[healiseu]

Faulty package installation was corrected by running again the quick start commands
Then I modified /etc/clickhouse-server/config.xml again and
BINGO
The clickhouse server started ok and the location of clickhouse/data changed to my SSD
There is one thing left. How to bring back the tables from the old server.
I kept a copy of the old data folder but I am not sure if I can restore them this way ?
For example I tried
ATTACH TABLE Floats
and I got back an exception
Exception: There is no metadata file for table Floats.
OK I created/copied the needed metadata at clickhouse/medatada and repeated the ATTACH command. That worked but I got back
0 rows in set
OK I have just discovered the "Recovery After Complete Data Loss" part of the documentation that explains how to do this
The easy way is to run the command
sudo -u clickhouse touch /var/lib/clickhouse/flags/force_restore_data

That's it, hope this hands-on failure/recovery experience of mine will be helpful for others.

[healiseu]

More about uninstalling and installing again clickhouse

I noticed that
apt purge clickhouse-server
apt purge clickhouse-client
are not enough you need also
apt autoremove
But even with these in place there is leftover of clickhouse from previous installation
So I also did several
rm -rf .../clickhouse*
and then installed again, otherwise it's quite possible that clickhouse-server will fail to start as it happened in my case.

[sevketozturk]

!!start the service after making changes to the config !!!
--note the ranking--
To install official packages add the Yandex repository in /etc/apt/sources.list or in a separate,
/etc/apt/sources.list.d/clickhouse.list file:
1- deb http://repo.yandex.ru/clickhouse/deb/stable/ main/
2- sudo apt-key adv --keyserver keyserver.ubuntu.com --recv E0C56BD4
3- sudo apt-get update
4- sudo apt-get install clickhouse-client clickhouse-server
5- start the service after making changes to the config (if necessary)
6- choown clickhouse:clickhouse clickhouse/
7- sudo -R chmod 755 clickhouse/
8- sudo service clickhouse-server start
done :)

[lexerom]

Also had such error. In logs I had: Application: Exception: Failed to merge config with '/etc/clickhouse-server/users.d/default-password.xml': SAXParseException: Invalid token in '/etc/clickhouse-server/users.d/default-password.xml', line 1 column 38.

It was easy to fix - I updated the password for default user - set it without special symbols, just 0-9a-Z.

[alexey-milovidov]

@lexerom You can also use sha256 password.

[depyronick]

I was deleted the error log file manually with user ubuntu and created it myself, so the permission got changed to ubuntu. /var/log/clickhouse and the files inside it should be chowned to clickhouse:clickhouse. That fixed the error.

[kreuzerkrieg]

Odd. I have the same error after the config was changed and log level node was left empty. syslog has a correct message:

Sep 8 09:11:00 ip-172-16-8-5 clickhouse-server[3774]: Poco::Exception. Code: 1000, e.code() = 0, e.displayText() = Invalid argument: Not a valid log level, Stack trace (when copying this message, always include the lines below):

No idea how it translates into status=232/ADDRESS_FAMILIES