From 71d3fb483035816cad4f495c112bd3235d96d897 Mon Sep 17 00:00:00 2001 From: Daiki Sekihata Date: Mon, 2 Feb 2026 21:38:30 +0100 Subject: [PATCH 1/2] Add troubleshooting section for shared memory errors Added troubleshooting tip for shared memory segment size issue. --- docs/troubleshooting/README.md | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/docs/troubleshooting/README.md b/docs/troubleshooting/README.md index 261d25b6..1155e6a5 100644 --- a/docs/troubleshooting/README.md +++ b/docs/troubleshooting/README.md @@ -180,6 +180,11 @@ Error in : Failed to connect to any server! Giving up - You should be greeted with `Welcome to the ALICE GRID`. - Exit with `exit`. +### shmem: could not create a message of size ... + +Add `--shm-segment-size 1000000000` (or larger number) to your o2 command. +It is due to shared memory segment that is probably outdated for the size of the intermediate tables you have. It is actually set as 90% of system's max VSIZE allowance set in limits, or 2GB if the limits cannot be found, which is probably what happens in your case. + ## Graphical issue with high DPI monitors When using monitors with high DPI the labels and text in TBrowser might be too small, especially when fractional scaling is enabled in multiple From 36bae4739ac010315eaede74ce8e227ffc15ced6 Mon Sep 17 00:00:00 2001 From: Daiki Sekihata Date: Mon, 2 Feb 2026 21:39:11 +0100 Subject: [PATCH 2/2] Update FAQ by removing outdated local machines section Removed section about local machines and shared memory issues. --- docs/faq/README.md | 6 ------ 1 file changed, 6 deletions(-) diff --git a/docs/faq/README.md b/docs/faq/README.md index f3006697..0fb467ce 100644 --- a/docs/faq/README.md +++ b/docs/faq/README.md @@ -38,9 +38,3 @@ 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. - -## Your local machines -### `shmem: could not create a message of size ...` - -Add `--shm-segment-size 1000000000` (or larger number) to your o2 command. -It is due to shared memory segment that is probably outdated for the size of the intermediate tables you have. It is actually set as 90% of system's max VSIZE allowance set in limits, or 2GB if the limits cannot be found, which is probably what happens in your case.