NETGEAR is aware of a growing number of phone and online scams. To learn how to stay safe click here.
ReadyDR with 10GB connection requires throttling
I was running out of room on my 6-drive RN426 backup unit (to my 8-drive RN628X main unit) and decided to swap in my "spare" RN528X and then add another drive. The 426 is just 1GB Ethernet while the 528X and 628X have 10GB and I have a 10GB switch, so faster backups would be an added bonus. Suddenly, my ReadyDR backup (used for my one share with Veracrypt files on it) started failing with a "broken pipe" error on a different file each time I ran it. At first, I thought something was wrong with the Ethernet connection on the 528X, so I swapped ports. That didn't help. I then went back to the 426 and all was well. But I was confused because my other (rsync) backups were working just fine on the 528 and with improved speed, and many of those shares are larger and have more churn than the one with the ReadyDR backup. I even did an rsync backup of that share, and it worked fine -- backing up the whole thing new, not incrementally. Long story short, I found an old article here in the forum that listed slowing it down as a possible fix to this type error (ReadyDR has a bandwidth limit option), so I tried that and it worked. Up to the point it quit, the reported speed before throttling was around 150MB/sec, so I first tried throttling to 75 and it went a little farther (and much longer because of the speed), but still failed. I had expected 75 to work and then try moving up incrementally, but it turns out I had to go down to 50 for it to reliably complete the approximately 1TB share backup. I'm unsure why ReadyDR needs so much throttling -- TOP shows there is plenty of CPU throughput available on both systems. My best guess is that BTRFS (which manages ReadyDR, which is Netgear's implementation of BTRFS send/receive) and/or MDADM RAID (RAID6 on the main one, RAID5 on the backup) just can't handle the required disk I/O at faster transfer speeds. If anyone knows of another possible fix, I'm all ears. But I mostly posted this in case anyone else is looking for a solution to this error. With the infrequent and then small changes in the affected share, I can live with the slower speed. Turning off quotas was also listed as a possible fix, but I didn't try that. I need to keep track of my drive usage. As an aside, I plan to keep the RN528 as RAID5 since it's just a backup and only on for the backups. Otherwise, I'd need to add two drives to get expansion since OS6 XRAID automatically switches to RAID6 on the 7th drive. See How to "fight" XRAID and win | NETGEAR Communities on how to do that (especially the second post).
