Closed dhedberg closed 7 months ago
Hello @dhedberg
did you update openvas
as well? or only ospd-openvas
?
Hello,
We're running the community containers, deployed in kubernetes with a setup adapted from the provided docker-compose example. The deploy is configured with these images:
image: docker.io/greenbone/cert-bund-data
image: docker.io/greenbone/data-objects
image: docker.io/greenbone/dfn-cert-data
image: docker.io/greenbone/gpg-data
image: docker.io/greenbone/gsa:stable
image: docker.io/greenbone/gvmd:stable
image: docker.io/greenbone/gvm-tools:latest
image: docker.io/greenbone/notus-data
image: docker.io/greenbone/notus-scanner:stable
image: docker.io/greenbone/ospd-openvas:stable
image: docker.io/greenbone/pg-gvm:stable
image: docker.io/greenbone/report-formats
image: docker.io/greenbone/scap-data
image: docker.io/greenbone/vulnerability-tests
It's restarted once per day and the pullpolicy is set to always for all containers, so we should always be running the latest images available for the tags. The scans that are broken haven't been working since the restart last Thursday (that is, around 2024-03-14 18:00 CET).
I have just run a scan against a target with 51 hosts. I excluded six hosts. Other hosts were dead. The scan finished successfully. The GUI shows the status as Done
and the scanner logs are the followings:
openvas.log
libgvm boreas:MESSAGE:2024-03-19 12h49.44 utc:2455155:02d84c10-6d29-4722-981c-bdc3851dfbc2: Alive scan 02d84c10-6d29-4722-981c-bdc3851dfbc2 finished in 5 seconds: 10 alive hosts of 45.
sd main:MESSAGE:2024-03-19 12h49.46 utc:2455155:02d84c10-6d29-4722-981c-bdc3851dfbc2: Vulnerability scan 02d84c10-6d29-4722-981c-bdc3851dfbc2 finished in 7 seconds: 10 alive hosts of 45
ospd-openvas.log
OSPD[$PID] 2024-03-19 12:49:47,764: INFO: (ospd.ospd) 02d84c10-6d29-4722-981c-bdc3851dfbc2: Scan finished.
OSPD[$PID] 2024-03-19 12:49:47,765: DEBUG: (ospd.ospd) 02d84c10-6d29-4722-981c-bdc3851dfbc2: Current scan progress: 100,
OSPD[$PID] 2024-03-19 12:49:47,765: DEBUG: (root) 02d84c10-6d29-4722-981c-bdc3851dfbc2: Current progress:
{'count_alive': 10,
'count_dead': 35,
'count_excluded': 6,
'count_total': 45,
'current_hosts': {},
'overall': 100}
I build from sources. Openvas is included in the ospd-openvas container. Could you check the openvas version inside the container or in the logs?
openvas
logs should show something like
sd main:MESSAGE:2024-03-19 12h49.35 utc:2455155:02d84c10-6d29-4722-981c-bdc3851dfbc2: openvas 23.0.1~git-ede9dcc0-http2 (GIT revision ~git-ede9dcc0-http2) started
I get this for openvas -V
OpenVAS 22.7.9
gvm-libs 22.4.1~dev1
Most new code since 2005: (C) 2022 Greenbone Networks GmbH
Nessus origin: (C) 2004 Renaud Deraison <deraison@nessus.org>
License GPLv2: GNU GPL version 2
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.
.. and this for ospd-openvas --version
OSP Server for openvas: 22.7.0
OSP: 22.7.0
OSPd OpenVAS: 22.7.0
Copyright (C) 2014-2021 Greenbone AG
License GPLv2+: GNU GPL version 2 or later
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.
Indeed it is an outdated openvas version. We will re build the images. Thank you very much for reporting this.
New image is available and it should include the last openvas version. I leave this issue open. Please check the new image and and close the issue if it works as expected. Thanks again.
It seems to working fine again now, thank you for the quick response!
You are welcome!
Since updating to the latest version with the stable tag last week (the 14th), we have some scans that now consistently end up in state interrupted instead of finishing.
The symptoms appear to be more or less exactly those in #951, for example:
For this scan, the target that fails looks basically like this:
When I do a test run now I appear to get 14 hosts alive and included in the range, and I note that
14/(14+8) = 0.636..
, so not accounting for the excluded hosts would appear to roughly explain the progress not being 100%. This may be a coincidence, I have not looked too closely at the changes.