You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Darshan has a finite table in memory to hold mounted file systems and a finite limit on the number of characters it will use to hold the table in the log file.
If more file systems are mounted then will fit in either location, then we should prioritize based on file system type to try to record the most interesting file systems. For example, we could defer NFS file systems to the end in order to prioritize parallel file systems over automounted home or project directories.
The text was updated successfully, but these errors were encountered:
For 2.2.9 we'll just take the initial step of iterating through mounted file systems in two passes, making sure to store the non-nfs ones first if there are too many to fit in the available header space.
In GitLab by @shanedsnyder on Sep 24, 2015, 16:25
Darshan has a finite table in memory to hold mounted file systems and a finite limit on the number of characters it will use to hold the table in the log file.
If more file systems are mounted then will fit in either location, then we should prioritize based on file system type to try to record the most interesting file systems. For example, we could defer NFS file systems to the end in order to prioritize parallel file systems over automounted home or project directories.
The text was updated successfully, but these errors were encountered: