Constant disk noise - anyway to check disk activity? (II)

[lolohz]

Hello
I see that TOS 4.2.09-2101251735 mounts /tmp in memory, but the disks that contains the O.S. are working all the time, writing every 5 seconds.

[admin@TerraNAS ~]$ cd /
[admin@TerraNAS /]$ df -T /tmp
Filesystem Type 1K-blocks Used Available Use% Mounted on
tmpfs tmpfs 1962864 1088 1961776 0% /tmp

I was a while spying the changed files on them and I discovered than many processes are writing logs and locks outside /tmp folder.

[admin@TerraNAS /]$ find . -not -path "./dev/*" -not -path "./proc/*" -mtime -0.001 -exec ls -d -la {} \; 2>&1 |grep -v find
-rw-r--r-- 1 root root 577040 Mar 31 02:24 ./opt/var/log/nginx/access.log
-rw-r--r-- 1 root root 30104 Mar 31 02:24 ./opt/var/log/crond.log
drwxrwxrwt 14 root root 840 Mar 31 02:24 ./tmp
-rw-r--r-- 1 root root 485 Mar 31 02:24 ./tmp/.tmp.tran.info
-rw-r--r-- 1 root root 485 Mar 31 02:24 ./tmp/.tmp.rec.info
drwxr-xr-x 2 root root 40 Mar 31 02:24 ./tmp/cron.g6aJ3O

[admin@TerraNAS /]$ find . -not -path "./dev/*" -not -path "./proc/*" -mtime -0.001 -exec ls -d -la {} \; 2>&1 |grep -v find
-rw-r--r-- 1 root root 2735 Mar 31 02:40 ./opt/var/log/samba/log.smbd
-rw-r--r-- 1 root root 97079 Mar 31 02:40 ./opt/var/log/messages
-rw-r--r-- 1 root root 32448 Mar 31 02:41 ./opt/var/log/crond.log
drwxr-xr-x 2 root root 40 Mar 31 02:41 ./tmp/cron.g6aJ3O
-rw-rw-rw- 1 root root 7077 Mar 31 02:39 ./tmp/sys_log
-rwxr-xr-x 1 root root 111 Mar 31 02:40 ./tmp/regcloud.request
-rwxr-xr-x 1 root root 2 Mar 31 02:40 ./tmp/regcloud.log

[admin@TerraNAS /]$ find . -not -path "./dev/*" -not -path "./proc/*" -mtime -0.001 -exec ls -d -la {} \; 2>&1 |grep -v find
drwxr-xr-x 7 root root 4096 Mar 31 02:50 ./etc/base
-rw-r--r-- 1 root root 3200 Mar 31 02:50 ./opt/var/log/user.log
-rw-r--r-- 1 root root 97132 Mar 31 02:50 ./opt/var/log/messages
-rw-r--r-- 1 root root 33694 Mar 31 02:50 ./opt/var/log/crond.log
-rw-r--r-- 1 root root 0 Mar 31 02:50 ./opt/var/lock/standby.lock
drwxr-xr-x 2 root root 140 Mar 31 02:50 ./opt/var/lock/samba/msg.lock
drwx------ 2 root root 140 Mar 31 02:50 ./opt/var/lib/samba/private/msg.sock
drwxr-xr-x 2 root root 40 Mar 31 02:50 ./tmp/cron.g6aJ3O
-rw-rw-rw- 1 root root 7433 Mar 31 02:49 ./tmp/sys_log
-rwxr-xr-x 1 root root 111 Mar 31 02:50 ./tmp/regcloud.request
-rwxr-xr-x 1 root root 2 Mar 31 02:50 ./tmp/regcloud.log

Would be possible please you to give us an option in TOS that allow us that all logs and locks were written in /tmp (notifying that you can lose relevant log data if the system is turned off)?

Would be possible please you to review next cron that is executed every minute to check if it is possible to change the target folder to /tmp?

[admin@TerraNAS /]$ ls -la /var/nas/everyminuteexec
-rwxr-xr-x 1 root root 793 Mar 30 18:35 /var/nas/everyminuteexec
[admin@TerraNAS /]$ cat /var/nas/everyminuteexec
#!/bin/bash
#fmt as $1:date $2:command $3....
echo [`date`] exec $0 >> /var/log/crond.log

wake_up_now() {
echo "aabbccddee" > /etc/base/.sync
sync
rm -f /etc/base/.sync
}

standby=0
config_file=/tmp/TOS_CONFIG/standby.conf

if [ -e $config_file ]; then
buffer=`awk -F= '/standby/ {print $2}' $config_file`
[ ! -z "$buffer" ] && standby=$buffer
fi

if [ "$standby" = "0" ]; then
locker=/var/lock/standby.lock
if [ ! -e $locker ]; then
touch $locker
else
mtime=`stat -c "%Z" $locker`
now=`date +%s`
let diff=$now-$mtime
# wake up after 10 mins ...
if [ $diff -le 600 ]; then
exit 1
fi
logger -t "TOS_STANDBY" "Wake up now..."
touch $locker
wake_up_now
fi
fi

exit 0

Could you please check some system events that should be avoided in order to not to write to many log entries like next one?

[admin@TerraNAS /]$ tail ./opt/var/log/messages
Mar 31 02:39:01 TerraNAS TOS_STANDBY: Wake up now...
Mar 31 02:40:01 TerraNAS smbd[3994]: [2021/03/31 02:40:01.463206, 0] ../source3/printing/print_standard.c:71(std_pcap_cache_reload)
Mar 31 02:40:01 TerraNAS smbd[3994]: Unable to open printcap file /etc/printcap for read!
Mar 31 02:50:01 TerraNAS TOS_STANDBY: Wake up now...
Mar 31 02:52:31 TerraNAS smbd[3994]: [2021/03/31 02:52:31.516029, 0] ../source3/printing/print_standard.c:71(std_pcap_cache_reload)
Mar 31 02:52:31 TerraNAS smbd[3994]: Unable to open printcap file /etc/printcap for read!
Mar 31 02:57:06 TerraNAS RESTART_SERVICE: 80 is empty...
Mar 31 02:57:09 TerraNAS RESTART_SERVICE: nasips total process is [0]...
Mar 31 02:57:40 TerraNAS last message repeated 10 times
Mar 31 02:58:10 TerraNAS last message repeated 10 times

Please, feel free to ask me any question by email, as I don't often read the forums.

Best regards.

[lolohz]

Hello
It is amazing, the disk stopped writing every 5 seconds.
I checked the folder /opt/var/log, where files were written, to see what changed and I realized that it is also in memory:

[root@TerraNAS /]# df -T /opt/var/log
Filesystem Type 1K-blocks Used Available Use% Mounted on
tmpfs tmpfs 1962868 5448 1957420 0% /opt/var


[root@TerraNAS /]# mount
/dev/md9 on / type ext4 (rw,noatime,nodiratime,commit=60,data=writeback)
devtmpfs on /dev type devtmpfs (rw,relatime,size=1947472k,nr_inodes=486868,mode=755)
proc on /proc type proc (rw,relatime)
devpts on /dev/pts type devpts (rw,relatime,gid=5,mode=620,ptmxmode=666)
tmpfs on /dev/shm type tmpfs (rw,relatime,mode=777)
tmpfs on /tmp type tmpfs (rw,relatime)
tmpfs on /run type tmpfs (rw,nosuid,nodev,relatime,mode=755)
sysfs on /sys type sysfs (rw,relatime)
tmpfs on /opt/var type tmpfs (rw,relatime)
/dev/mapper/vg0-lv0 on /mnt/md0 type ext4 (rw,noatime,block_validity,delalloc,barrier,user_xattr,acl,jqfmt=vfsv1,usrjquota=quota.user,grpjquota=quota.group)
nfsd on /proc/fs/nfsd type nfsd (rw,relatime)


So I suspect that the origin of such activity was a smb connection or some kind of hdd/raid initial setup.
Anyway I 'll watch over some days to confirm if it was a transient condition.
Best regards