[NPMCLI-76] Investigate how newcomers can easily overcome the open files issue Created: 19/Jan/17  Updated: 12/Apr/17

Status: Open
Project: Magnolia CLI
Component/s: None
Affects Version/s: None
Fix Version/s: None

Type: Task Priority: Neutral
Reporter: Robert Kowalski Assignee: Unassigned
Resolution: Unresolved Votes: 0
Labels: blocked
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
relation
is related to MGNLCE-60 Remove "too many open files" startup ... Closed
Template:
Acceptance criteria:
Empty
Task DoR:
Empty
Date of First Response:

 Description   

For some users the tomcat won't start because their system runs out of file descriptors.

It would be nice to provide them with an easy way to fix the issue. This could be a command or an instruction how to solve the issue.

Reference:

https://documentation.magnolia-cms.com/display/DOCS/_Known+issues



 Comments   
Comment by Robert Kowalski [ 23/Jan/17 ]

Important: Maxime told me that they want to remove that check as it's not needed anymore

Comment by Christopher Zimmermann [ 23/Jan/17 ]

blocked for reason stated above.
Address the test described in linked MGNLCE-60 before proceeding.

Comment by Robert Kowalski [ 23/Jan/17 ]

from the planning session: the errors can manifest in many different way as file descriptors are used for many different things by magnolia (e.g. http requests for asset files, database requests and so on).

every os comes with different limits and has different services already running which consume FD.

resolution: keep the warning, but improve how users get to know the issue and help them to fix it.

Comment by Robert Kowalski [ 11/Apr/17 ]

mmichel confirmed today: inside a docker container its possible without making changes to the host system downside: depends on docker

Comment by Maxime Michel [ 12/Apr/17 ]

Subtle nuance: I confirmed that it seems to work With a host with a limit of about 4000, I could indeed hit about 100'000 descriptors in the container. However, I don't know enough about kernel internals to understand how that is possible. It would be interesting to test a Docker container with two simultaneous EE instances running - with a low limit this used to crash badly on my host.

Generated at Mon Feb 12 04:46:17 CET 2024 using Jira 9.4.2#940002-sha1:46d1a51de284217efdcb32434eab47a99af2938b.