Pgscv_problem

[gmaharram]

After starting pgscv, although pgscv's status is active, but it gives following error:

 **service":"pgscv","time":"2022-06-23T15:33:32+04:00","message":"system/sysinfo collector failed; invalid input, '': wrong number of values"}**

Also, i connect my database to prometheus and grafana, and in every 3 minute it returns no data, and after 30 seconds i am able to see data means i loose some data. So, I changed restart time to 1 sec, and now after 12 minutes for 2minutes there are not any data repeatedly.

OS: Centos 7
pgSCV Version: 0.7.5
Postgres: 11
Postgres and pgSCV are running on the same host

pgscv.yml:

listen_address: ip_address:9890
services:
  "postgres:5432":
    service_type: "postgres"
    conninfo: "postgres://postgres:somepassword@ip_address:5432/postgres"
defaults:
    postgres_username: "pgscv"
    postgres_password: "somepassword"
no_track_mode: false

[lesovsky]

Seems pgscv fails to read /etc/os-release file. Check that file exists and show its content pls

[lesovsky]

in every 3 minute it returns no data, and after 30 seconds i am able to see data means i loose some data.

is there any other errors, except this "system/sysinfo collector failed; invalid input, '': wrong number of values" ? If no, you should check logs of prometheus/grafana.

[gmaharram]

etc/os-release gives this output

NAME="CentOS Linux"
VERSION="7 (Core)"
ID="centos"
ID_LIKE="rhel fedora"
VERSION_ID="7"
PRETTY_NAME="CentOS Linux 7 (Core)"
ANSI_COLOR="0;31"
CPE_NAME="cpe:/o:centos:centos:7"
HOME_URL="https://www.centos.org/"
BUG_REPORT_URL="https://bugs.centos.org/"

CENTOS_MANTISBT_PROJECT="CentOS-7"
CENTOS_MANTISBT_PROJECT_VERSION="7"
REDHAT_SUPPORT_PRODUCT="centos"
REDHAT_SUPPORT_PRODUCT_VERSION="7"

[gmaharram]

{"level":"error","service":"pgscv","time":"2022-06-23T16:14:32+04:00","message":"system/sysinfo collector failed; invalid input, '': wrong number of values"}
{"level":"error","service":"pgscv","time":"2022-06-23T16:15:02+04:00","message":"system/sysinfo collector failed; invalid input, '': wrong number of values"}
{"level":"warn","service":"pgscv","time":"2022-06-23T16:15:32+04:00","message":"get model for fd0 failed: open /sys/block/fd0/device/model: no such file or directory; skip"}

[lesovsky]

Hm, it fails when reading empty line in this file. Will fix in next release.

As a workaround you can remove empty line in the os-release file.

[gmaharram]

Thank you for answering, also during no data i mean for 30 seconds curl -s http://127.0.0.1:9890/metrics | grep -c ^postgres this command gives 0

[lesovsky]

If you see no any other errors in pgSCV logs related to postgres, checkout Postgres logs and logs of other components.

[gmaharram]

I removed the empty line, but error is the same

[lesovsky]

When I adding empty line, i see that error. When remove empty line, error disappear. Also I have a unit test which helps me to reproduce that behaviour.

[gmaharram]

Thank you it helped me a lot, but I got this warning again
{"level":"warn","service":"pgscv","time":"2022-06-23T16:41:32+04:00","message":"get model for fd0 failed: open /sys/block/fd0/device/model: no such file or directory; skip"}

[lesovsky]

Messages with warning severity are not critical, you could ignore it or reduce logging level from info (default) to error and omit them.

[gmaharram]

There are not errors in PostgreSQL logs and pgscv, but I am not able to see some data again for 2 minutes continuously.

[lesovsky]

Show me the first lines of logs after pgscv is started (before the line with "message":"listen on http...")

[gmaharram]

Jun 28 17:22:13 testdb systemd[1]: Started pgSCV is the Weaponry platform agent for PostgreSQL ecosystem.
Jun 28 17:22:14 testdb pgscv[19263]: {"level":"info","service":"pgscv","time":"2022-06-28T17:22:14+04:00","message":"read configuration from /etc/pgscv.yaml"}
Jun 28 17:22:14 testdb pgscv[19263]: {"level":"info","service":"pgscv","time":"2022-06-28T17:22:14+04:00","message":"no-track disabled, for details check the documentation about 'no_track_mode' option."}
Jun 28 17:22:14 testdb pgscv[19263]: {"level":"info","service":"pgscv","time":"2022-06-28T17:22:14+04:00","message":"registered new service [system:0]"}
Jun 28 17:22:14 testdb pgscv[19263]: {"level":"info","service":"pgscv","time":"2022-06-28T17:22:14+04:00","message":"registered new service [postgres:5432]"}

[lesovsky]

Ok, as i can see "postgres" service is registered. Now run pgscv with "--log-level=debug". It will produce a lot of debug messages. In another console run "curl -s http://127.0.0.1:9890/metrics | grep -c ^postgres" several times. When command returns 0 metrics, inspect debug logs for any suspicious messages.

[gmaharram]

Thank you for your help, I think it is because of VM