Problem with "Archive size" in health monitor
Hello,
I have a problem with correctly displaying the archive size in the health monitor.
Piler version: piler 1.4.4, build 1001 (It’s a fresh install).
I use manticore and realtime indexer.
Piler doesn't count the size of the archive, received messages and duplicates when I use pilerimport.
The archive size is only updated when archiving via SMTP.
I checked “counter” table in database, pilerimport doesn’t update records in the database.
Comments (5)
-
repo owner -
repo owner - changed status to open
-
reporter Hello,
I'm confused because I don't know how it works. This is a fresh environment, retention is set to 5 years.
Currently, I'm archiving mail via always bcc and older message by pilerimport.
Counter table:
select * from counter; +---------+-------+-----------+--------+---------------+---------------+ | rcvd | virus | duplicate | ignore | size | stored_size | +---------+-------+-----------+--------+---------------+---------------+ | 7663098 | 0 | 3101567 | 0 | 3440766526989 | 1274292971184 | +---------+-------+-----------+--------+---------------+---------------+
RCVD items from counter isn’t equal to number of record from metadata:
MariaDB [piler]> select count(*) from metadata; +----------+ | count(*) | +----------+ | 19083807 | +----------+
Sum size from metadata:
MariaDB [piler]> select sum(size) AS TotalItemsOrdered from metadata; +-------------------+ | TotalItemsOrdered | +-------------------+ | 5219603635348 |
Size from counter isn’t equal to size from metadata.
Stored_size from counter isn’t equal to size from df -h command (Stored_size ~ 1.27TB) :
root@z-mailpiler-n:/var/piler/migracja# df -h Filesystem Size Used Avail Use% Mounted on tmpfs 1,6G 1,5M 1,6G 1% /run /dev/mapper/ubuntu--vg-ubuntu--lv 38G 17G 19G 48% / tmpfs 7,9G 40K 7,9G 1% /dev/shm tmpfs 5,0M 0 5,0M 0% /run/lock /dev/sda2 2,0G 191M 1,6G 11% /boot /dev/mapper/ubuntu--vg-backup 10T 2,0T 7,5T 21% /var/piler
I don't understand what I'm doing wrong
-
repo owner Try the following. Stop the piler daemons, then run “select * from counter”. Next, import a single email, and run the sql query again, then compare the results.
The rcvd, size and stored_size columns should increase.
-
repo owner - changed status to closed
No further input on the matter.
- Log in to comment
I can’t reproduce the issue. I’ve just imported a few emails, and the counters are updated just fine.