GKE build extremely slow

no flag

I have a build that typically requires 2.5 hours to redeploy now been running for the last 20 hours.

There are some disk operations on it. In particular, each NFS read/write appears 10,000 times slower than normal.

When looking at the Google Console Status for Google Cloud, I see that Google had many breaks in their infrastructure over the last couple of days, but I cannot relate any of them to my current issue and all services are now green.

Is there an online form to test the throughput of one's GCP storage? This could help get some reliable metrics that are subjective rather than to continue to be in the subjective.

Any guidance would be the most welcome.

Arden Smith avatar
pe flag
As per your issue description, this issue is related to a GKE. Could you describe the kind of nodes and application behind this?
John Mahowald avatar
cn flag
Regarding NFS, is this managed Filestore or something you run? What does your storage workload look like, average I/O size, typical and peak IOPS? What do you observe for latency, in particular median and maybe 99th percentile? I believe you that its being slow, but 10,000x slower doesn't convey what you expect out of this storage, and what is is delivering.

Post an answer

Most people don’t grasp that asking a lot of questions unlocks learning and improves interpersonal bonding. In Alison’s studies, for example, though people could accurately recall how many questions had been asked in their conversations, they didn’t intuit the link between questions and liking. Across four studies, in which participants were engaged in conversations themselves or read transcripts of others’ conversations, people tended not to realize that question asking would influence—or had influenced—the level of amity between the conversationalists.