NETGEAR is aware of a growing number of phone and online scams. To learn how to stay safe click here.
Forum Discussion
systemd journalctl Logging
Hi all,
I am slowly getting used to some of the changes between Legacy models and OS6 whilst in the process of migrating from an aging Pro6 to a 526X.
One of the differences which I noticed fairly early on was the adoption of systemd / journalctl for logging. I actually prefer seperate log files and it's what I've been used to in my 15 years experience working with Linux. I have no doubt that leveraging a logging database probably scales superbly for a large scale environment but so far I've mainly found cons of using it on the ReadyNAS.
One of the major flaws I found today following a bunch of rsync data transfers yesterday is that all my previous logs have now been overwritten. Seemingly a single process has been able to fill the log and cause older entries to be purged and these were just normal file transfers albeit circa 500 GB. I'm sure there is a setting to increase the overwrite threshold but I have no intention of fiddling with settings via SSH as this can cause Netgear to deny support.
It may be that I have missed something fundamental here but personally I'd rather have individual log files.
7 Replies
Replies have been turned off for this discussion
Though I agree the OS space needs to be small, I am wondering if they would have been better off to put the logs in their own partition (sized large enough to give a longer lookback).
Were the rsyncs done with backup jobs or via ssh? I haven't seen much impact on log duration due to backup jobs.
Rsyncs were done via Backup jobs but from another NAS. This was the destination device, otherwise they would have been within seperate backup logs I'd assume.
The logs zip includes recent lines of the journal in systemd-journal.log in the logs zip file. If you SSH in you'll see that the journal can grow to several times that size. kernel.log for instance shows recent lines in the journal that are about the kernel so would go further back in time than entries in systemd-journal.log.
If you do encounter a problem it's best to look at the logs ASAP before they are rotated.
- Thanks. So how do I request from Cali the kernel entries like you reference? I'd probably pipe it to grey using kernel
You could do e.g.
# journalctl -a --no-pager --lines 5000 _TRANSPORT=kernel
NETGEAR Academy
Boost your skills with the Netgear Academy - Get trained, certified and stay ahead with the latest Netgear technology!
Join Us!
