I need to make a crontab to gzip a file named mh located on my desktop every 2 minutes in the same path. I tried
2 * * * * gzip home/Desktop/mh >> home/Desktop
But it is not working, any help is greatly appreciated.
There are several errors here.
The gzip command should be simply gzip home/Desktop/mh. Remove the >> and everything afterwards.
Your current cron will only run on the second minute of every hour. Instead you want */30 * * * * ... to run 30 times per hour.
Note that gzip is "destructive" in the sense that your original file (mh) will disappear after each gzip. That would be bad if some other process is trying to write to it continually...
If you want to keep the content of mh and just update mh.gz from it periodically, you want to do
*/30 * * * * gzip < /home/Desktop/mh > /home/Desktop/mh.gz
Related
I tried to use crontab to execute my py file everyday, but it can only create empty log file
0 8 * * * /usr/local/bin/python3 /Users/UserName/Downloads/Crawling_1.py > /Users/UserName/Downloads/log.log
Then I tried to use SHELL file to execute a simple demand, if I put log file settings inside .sh file, no log file was created. Similarly, the crontab did not execute when I put python3 demand inside the SHELL file.
echo 1 > /Users/UserName/Downloads/new_log.log > /Users/UserName/Downloads/log.log
But if I directly run echo in Crontab, it can work out perfectly.
* * * * * echo 1 > /Users/UserName/Downloads/new_log.log
Does anyone know why this is happening? Thank you so much.
Try it with >>:
0 8 * * * /usr/local/bin/python3 /Users/UserName/Downloads/Crawling_1.py >> /Users/UserName/Downloads/log.log
then with >> it will create a file if it doesn't exist. If the file existe, it will be appended to the end of the file.
With > the whole file will replace it, if the file exist. If the file not exist*, nothing happens.
You can do the command also with >, but be sure, that the .log-file, where you will write inside, exist!
I have a crontab job set up which records the speed of my network every two minutes:
*/2 * * * * /usr/local/bin/speedtest >> ~/Documents/speedtest.log
The output is saved to speedtest.log:
Testing download speed................................................................................
Download: 60.10 Mbit/s
Is there a way to add a timestamp (the format given is an example only; the actual format is irrelevant) to each entry to achieve something like this?
20200125221000: Testing download speed................................................................
20200125221042: Download: 60.10 Mbit/s
If so, how?
You can do:
*/2 * * * * echo "$(/usr/local/bin/speedtest) $(date)" >> ~/Documents/speedtest.log
I have several scripts that are run every 3 minutes and schedule looks like this:
*/3 * * * * /some/script1.php
*/3 * * * * /some/script2.php
*/3 * * * * /some/script3.php
I suppose that these scripts run at the same time, but I wish that these scripts run every 3 minute but not in the same time. Tell me please how can I reach this.
You can't reach that with the lines you are having, those will allways be running at the same times. However, you can simply create one "master" script that gets called via cron and then calls the scripts one after the other.
*/3 * * * * /usr/bin/php /some/masterscript.php
masterscript.php:
<?php
exec('/usr/bin/php /some/script1.php');
exec('/usr/bin/php /some/script2.php');
exec('/usr/bin/php /some/script3.php');
?>
EDIT:
Depending on your server's setup - install node.js. There's a cron package you can set for every second. Maybe this can help..
You can but not with that syntax, instead you should use this kind of syntax:
1,4,7,10,13,16,etc... * * * * /some/script1.php
2,5,8,11,14,17,etc... * * * * /some/script2.php
and so on....
If you just want each script to execute in turn, with the second not starting until the first has finished, and so forth, just put them all in a single cron command.
cron invokes each command by passing the command string to /bin/sh -- and the shell can very easily invoke several commands in sequence.
*/3 * * * * /some/script1.php ; /some/script2.php ; /some/script3.php
do you think a couple of seconds between scripts run could be enough ?
what about a command like this ?
*/3 * * * * echo "<?php echo 'Start ...';sleep(2);echo Go; ?>"|php /some/script1.php
*/3 * * * * echo "<?php echo 'Start ...';sleep(4);echo Go; ?>"|php /some/script2.php
*/3 * * * * echo "<?php echo 'Start ...';sleep(6);echo Go; ?>"|php /some/script3.php
You could also substitute fixed waiting time with random waiting time.
Instead of sleep(2) try a generic sleep(rand(1,10)).
I hope this could be useful
I have a crontab running like:
*/15 * * * 4,5 /apps/ins/sid/compare_stats 2>> /apps/ins/sid/compare_stats.err
Everything working as expected. the only thing is I want my error logs to generate in the compare_stats.err file like this:
Jul 3 14:45:04 <error text>
which means I just want to add a date along with this. Is there any way to do it by modifying the crontab entry ( without really making any change in my script) ?
Thanks in advance.
Use the ts command which is part of the moreutils package. E.g.:
*/15 * * * 4,5 /apps/ins/sid/compare_stats | ts '[%Y-%m-%d %H:%M:%S]' 2>> /apps/ins/sid/compare_stats.err
This will prepend the timestamp to every line of the output and save it into your log.
I'm running a slice of ubuntu hardy.
I've installed sphinx, and I would like to run the sphinx indexer every x minutes.
What is the best way to go about doing this?
The standard Unix approach is cron, so you could for example edit /etc/crontab and add a line like
*/5 * * * * root sphynx [whatever other options you need]
which means
'every five minutes' (for the */5 part)
of every hour (the * in position 2)
of every day of the month (the * in position 3)
of every month (the * in position 4)
of every day of the week (the final * in position 5)
Another example: '4 5 * * 6' amounts to 'at 5:04am (four minutes after five) on every Saturday (day of the week is 6).
You may need to or want to switch the user from root to, say, www-data is sphynx runs as that, and you obviously need to adjust the arguments.
Lastly, look into the directories
$ ls -1d /etc/cron.*
/etc/cron.d
/etc/cron.daily
/etc/cron.hourly
/etc/cron.monthly
/etc/cron.weekly
for examples --- other packages put their jobs there (and this mechanism is more general, and newer, than direct editing of /etc/crontab.
Here is what I do to reindex and then restart the search daemon once a day.
* * /1 * * root cd /home/sphinx && bin/indexer --all --rotate && bin/searchd --stop && bin/searchd