Unfortunately, the severe download performance degradation is still ongoing. I opened a support ticket with OVH last week, but despite following up, we have not received a response or resolution from them. Since the issue is having a direct impact on CI workloads and is causing some tests to time out, I have started migrating download.ceph.com to a new OVH location where I have verified that both download and upload speeds are operating normally. There is approximately 4 TB of data to migrate. Around 2 TB has already been copied, with approximately 2 TB remaining. I hope to complete the data transfer by the end of today and then switch download.ceph.com to the new instance. I’ll send another update once the migration is complete and the service has been switched to the new instance. Thanks, Adam -- Adam Kraitman Systems Administrator Ceph Engineering IRC: akraitma
This turned out to be (probably) an unintentional DDoS by a handful of GoDaddy IP addresses that resolve to *.secureserver.net which I’ve blocked. Download speeds to the lab have increased from ~50 KB/s to 4.5MB/s. In addition to blocking those IPs, we’ll be migrating download.ceph.com to a newer OVH instance type that will be cheaper and give us higher bandwidth throughput limits. Migration date TBD. From: Adam Kraitman <sepia@ceph.io> Date: Tuesday, September 15, 2026 at 9:21 AM To: sepia@ceph.com <sepia@ceph.io>; csc@ceph.io <csc@ceph.io>; dev <dev@ceph.io> Subject: [EXTERNAL] [sepia] download.ceph.com performance issue This Message Is From an External Sender This message came from outside your organization. Report Suspicious<https://us-phishalarm-ewt.proofpoint.com/EWT/v1/AdhS1Rd-!-XFVHHVS8ifQW9UUzgRKfdGJYKRMenu_GKIumul3kVncvX_OvF4VFs7d4CylROuNRdbGC7RJpuc5eflzQUwKldqhM6Yibziphj3cL_FgL3xJmCT5URnBO8VCg83NmTo$> Unfortunately, the severe download performance degradation is still ongoing. I opened a support ticket with OVH last week, but despite following up, we have not received a response or resolution from them. Since the issue is having a direct impact on CI workloads and is causing some tests to time out, I have started migrating download.ceph.com<https://urldefense.proofpoint.com/v2/url?u=http-3A__download.ceph.com&d=DwMFaQ&c=BSDicqBQBDjDI9RkVyTcHQ&r=MArCdIscFQh1RTbzeh581PBKkdx0cRfUxMPW6xo5kg8&m=DMFoEA5bSKzEtJK9THuSAaFyFRPy8rd96EppyMsHQ5tKr5LER2y8MHvdnbCPgM2z&s=bY1T5zy2r0PlueR_Ika3x0DyTs2VEZPAI_5DuBvqb28&e=> to a new OVH location where I have verified that both download and upload speeds are operating normally. There is approximately 4 TB of data to migrate. Around 2 TB has already been copied, with approximately 2 TB remaining. I hope to complete the data transfer by the end of today and then switch download.ceph.com<https://urldefense.proofpoint.com/v2/url?u=http-3A__download.ceph.com&d=DwMFaQ&c=BSDicqBQBDjDI9RkVyTcHQ&r=MArCdIscFQh1RTbzeh581PBKkdx0cRfUxMPW6xo5kg8&m=DMFoEA5bSKzEtJK9THuSAaFyFRPy8rd96EppyMsHQ5tKr5LER2y8MHvdnbCPgM2z&s=bY1T5zy2r0PlueR_Ika3x0DyTs2VEZPAI_5DuBvqb28&e=> to the new instance. I’ll send another update once the migration is complete and the service has been switched to the new instance. Thanks, Adam -- Adam Kraitman Systems Administrator Ceph Engineering IRC: akraitma
participants (2)
-
Adam Kraitman
-
David Galloway