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
{{ message }}
This repository has been archived by the owner on Oct 22, 2019. It is now read-only.
This is a stub to track a PR that I hope to write in future. However, I don't know when, or how, so I'm creating an issue to track interest.
I'd like to implement a Filesystem adapter. There are cases where, in shared hosting, I cannot install additional software onto a machine, or the provisioning of such a software would be terribly complex. However, there is access to a semi permanent directory where I can store files associated with an application, such as in the case of Magento, the var directory. Thus, I would like to implement a filesystem adapter that allows reading current metric state and modification thereof from disk.
Although it might be expensive in terms of IO, I would probably write to disk every time a metric incremented rather than batching the metrics in memory and writing them during destruct. The logic is, in some cases destruct may not be reached during abnormal application shutdown, and metrics would be useful for detecting / diagnosing such cases.
If anyone has any thoughts on this, please comment.
Hallo;
This is a stub to track a PR that I hope to write in future. However, I don't know when, or how, so I'm creating an issue to track interest.
I'd like to implement a Filesystem adapter. There are cases where, in shared hosting, I cannot install additional software onto a machine, or the provisioning of such a software would be terribly complex. However, there is access to a semi permanent directory where I can store files associated with an application, such as in the case of Magento, the
var
directory. Thus, I would like to implement a filesystem adapter that allows reading current metric state and modification thereof from disk.Although it might be expensive in terms of IO, I would probably write to disk every time a metric incremented rather than batching the metrics in memory and writing them during destruct. The logic is, in some cases destruct may not be reached during abnormal application shutdown, and metrics would be useful for detecting / diagnosing such cases.
If anyone has any thoughts on this, please comment.
See https://github.com/littlemanco/magento-prometheus for implementation.
The text was updated successfully, but these errors were encountered: