Unable to fetch configuration - Unresolved software bug
I know this has been flagged previously in other threads however I thought I'd put a bit more focus on it.
We have a customer (large retailer) who has a fleet of MX64's. In the last 3 weeks we have had 5 occurances where "Unable to fetch configuration" was displayed on the dashboard. This co-incides with the MX going offline - at least from the dashboard's perspective.
Customer was running 14.39 software. Whilst a call to Meraki resolved the issue, they had to fix something in the backend concerning local storage on the MX, the customer is still impacted. Apparently Meraki fixed and flagged this behaviour as a rare occurance according to release notes, 14.35, thanks @PhilipDAth
"Resolved an issue that could result in devices failing to pull configuration updates from the Meraki Dashboard in rare circumstances"
So not only is the issue _not_ fixed, it certainly is _not_ a rare circumstance. Too boot the Meraki employee I spoke to informed me that this issue is fixed in 14.40 despite the rls notes not stating such. Was informed by same individual not all bug fixes are stated in the release notes. I took it on face value and requested the customer to upgrade to 14.40 to prevent the issue re-occuring, which they did only to find that another event of the issue occured.
Was informed by different Meraki support engineer that this bug is still present and it's still under investigation.
This does not sit well with my customer, especially coming into the Christmas period.
Have requested if Meraki can proactivily perform some action to prevent downtime during business hours but to no avail. Have also requested list of devices in which the storage space is critical in a effort to help identify which units will suffer the same fate, no avail.
Have also been informed that even the RC and BETA software don't have any fix for this issue.
These units were installed approximately 3 years as part of a large project and it's almost as if the internal storage has been filling up eversince and reached a critial threshold causing this anomoly, so I suspect that most units that are part of this fleet will have the same fate at some stage.
There is a detailed internal document regarding this issue however it's not public facing.
Have asked Meraki to reference the bug ID and a commitment that the issue has high visibility with L4 engineering. Hopefully my customer will swallow that instead of a RCA.