From ebb00fa3fc972482be880fc982c75fb412dbae5e Mon Sep 17 00:00:00 2001 From: Daiki Sekihata Date: Fri, 30 Jan 2026 15:03:56 +0100 Subject: [PATCH] Update FAQ with shmem error resolution Added troubleshooting tip for 'shmem' error in O2 command. --- docs/faq/README.md | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/docs/faq/README.md b/docs/faq/README.md index 4db1cac1..5e1be825 100644 --- a/docs/faq/README.md +++ b/docs/faq/README.md @@ -8,6 +8,8 @@ title: Frequently asked questions (FAQ) This document lists a number of questions that have been asked by the community. If you think you have a relevant question to add, please don't hesitate to contact docadmin and we will be happy to add it. +- If you see `shmem: could not create a message of size ...`, add --shm-segment-size 1000000000 (or larger number) to your o2 command. + WILL BE EXPANDED - WORK IN PROGRESS ## Task writing @@ -37,4 +39,4 @@ Hyperloop trains have a so-called express train feature. This feature is based o ### Why is it that my train test has a CPU warning but my wagon test was fine? -This usually happens in a situation where the wagon test (which runs on a single core) uses so much memory that it doesn't fit a single core job on the grid and therefore needs two cores for the train (more cores means a higher memory allowance). But if the devices in the wagon cannot be parallelised well over multiple cores, this leads to more wall time and a higher CPU usage as the cores will be underutilised. In this situation, one can either reduce the wagon memory consumption to fit into a single core or reduce the CPU consumption to fit the dataset. \ No newline at end of file +This usually happens in a situation where the wagon test (which runs on a single core) uses so much memory that it doesn't fit a single core job on the grid and therefore needs two cores for the train (more cores means a higher memory allowance). But if the devices in the wagon cannot be parallelised well over multiple cores, this leads to more wall time and a higher CPU usage as the cores will be underutilised. In this situation, one can either reduce the wagon memory consumption to fit into a single core or reduce the CPU consumption to fit the dataset.