Hi Antonin,

On 8/21/26 2:26 PM, Antonin Godard via lists.openembedded.org wrote:
Add a "core/yocto/monitor-disk-space" fragment to enable the
functionality previously provided by the default definition of
BB_DISKMON_DIRS in local.conf.sample. As the definition might not be
obvious to set, having a simple way of enabling the feature through a
fragment makes it more approachable.

Signed-off-by: Antonin Godard <[email protected]>
---
  meta/conf/fragments/yocto/monitor-disk-space.conf | 18 ++++++++++++++++++
  1 file changed, 18 insertions(+)

diff --git a/meta/conf/fragments/yocto/monitor-disk-space.conf 
b/meta/conf/fragments/yocto/monitor-disk-space.conf
new file mode 100644
index 0000000000..95ff54465f
--- /dev/null
+++ b/meta/conf/fragments/yocto/monitor-disk-space.conf
@@ -0,0 +1,18 @@
+BB_CONF_FRAGMENT_SUMMARY = "Monitor disk space during the build"
+BB_CONF_FRAGMENT_DESCRIPTION = "Monitor the disk space during the build. If \
+there is less that 1GB of space or less than 100K inodes in any key build \
+location (TMPDIR, DL_DIR, SSTATE_DIR), gracefully shutdown the build. If there 
\
+is less than 100MB or 1K inodes, perform a hard halt of the build. The reason \
+for this is that running completely out of space can corrupt files and damages 
\

Remove "The reason for this is that".

I can suggest also rewording after the "and":

and can make unrecoverable damages to the build.

which is lighter (to me) when reading.

+the build in ways which may not be easily recoverable. It's necessary to 
monitor \
+/tmp, if there is no space left the build will fail with very exotic errors."

"No space left" is implied by the context of the text above so I think it can be removed.

-very, exotic is already strong enough to me.

+
+BB_DISKMON_DIRS = "\> +    STOPTASKS,${TMPDIR},1G,100K \
+    STOPTASKS,${DL_DIR},1G,100K \
+    STOPTASKS,${SSTATE_DIR},1G,100K \
+    STOPTASKS,/tmp,100M,100K \

According to your commit log, this should be 1K instead? but I guess this comes straight from meta/conf/templates/default/local.conf.sample where the inconsistency already exists.

Cheers,
Quentin
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#243954): 
https://lists.openembedded.org/g/openembedded-core/message/243954
Mute This Topic: https://lists.openembedded.org/mt/120860362/21656
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

Reply via email to