Ready NAS 102, slow network speed (and saving data on reset)
RN102 (6.9.3) on Gigabit network using Lan Speed Test I'm getting 80/90 Mbps (not MB/s), similar with a known good cable connected directly. Same speed from different machines. Antivirus not enabled All lights are green on the various network adapters.
Disks (JBOD) Seagate Ironwold 6Tb - st6000VN0041
Seagate NAS 4TB - st4000VN000
USB - Seagate backup hub in NTFS - 6TB
Q1. Has anyone got any ideas of what could cause this before I factory reset it. To be fair it's been like this for a while and it's only now the annoying speed is an issue.
Q2. If I do a factory reset on the NAS I will loose all data from the main disks? But, assuming I disconnect it, the USB disk should be find to keep as is?
Q1. Has anyone got any ideas of what could cause this before I factory reset it. To be fair it's been like this for a while and it's only now the annoying speed is an issue.
Security software on the PCs, using disk encryption on the NAS, SMB strict sync and transport encryption might have a performance impact.
Q2. If I do a factory reset on the NAS I will loose all data from the main disks? But, assuming I disconnect it, the USB disk should be find to keep as is?
Re: Ready NAS 102, slow network speed (and saving data on reset)
Thanks for the reply. Much appreciated.
Just to confirm, no disk encryption. Not sure what strict sync is but will look.
I've got the logs but am not sure where I'm looking for ethernet stats please?
NAS performance tester 1.7 is better..... Running warmup... Running a 400MB file write on U: 5 times... Iteration 1: 32.52 MB/sec Iteration 2: 30.35 MB/sec Iteration 3: 28.51 MB/sec Iteration 4: 31.28 MB/sec Iteration 5: 30.60 MB/sec ----------------------------- Average (W): 30.65 MB/sec ----------------------------- Running a 400MB file read on U: 5 times...
Would still like to see it getting a bit higher ( real copying is about 10Mb) but much more reassuring. I've seen it should be capable of 70-80, 50 would do me.
Having looked a bit further I can see also a ton of errors in journalctl
Oct 15 23:08:35 NETNAS kernel: mvneta d0074000.ethernet eth0: bad rx status 0f830000 (overrun error), size=512 Oct 15 23:08:36 NETNAS kernel: mvneta d0074000.ethernet eth0: bad rx status 0f830000 (overrun error), size=1024 Oct 15 23:08:36 NETNAS kernel: mvneta d0074000.ethernet eth0: bad rx status 0d830000 (overrun error), size=16 Oct 15 23:08:36 NETNAS kernel: mvneta d0074000.ethernet eth0: bad rx status 0f830000 (overrun error), size=1024 Oct 15 23:08:37 NETNAS kernel: mvneta d0074000.ethernet eth0: bad rx status 0f830000 (overrun error), size=896 Oct 15 23:08:37 NETNAS kernel: mvneta d0074000.ethernet eth0: bad rx status 0f830000 (overrun error), size=1024 Oct 15 23:08:37 NETNAS kernel: mvneta d0074000.ethernet eth0: bad rx status 0f830000 (overrun error), size=1024
Another post seems to suggest turning on flow control for this, but that it should be a problem - could this be connected? Is that advice accurate?
Also seeing a lot of leafp2p errors but assume that's not connected/ for another day
Oct 15 22:49:56 NETNAS leafp2p[2178]: url https://peernetwork.netgear.com:443//peernetwork/services/LeafNetsConfigWebService Oct 15 22:49:56 NETNAS leafp2p[2178]: SOAP 1.1 fault: SOAP-ENV:Client [no subcode] Oct 15 22:49:56 NETNAS leafp2p[2178]: "Name or service not known" Oct 15 22:49:56 NETNAS leafp2p[2178]: Detail: getaddrinfo failed in tcp_connect()
Re: Ready NAS 102, slow network speed (and saving data on reset)
But keep in mind resetting the NAS with a different drive is only going to let you test to see if the OS is the issue, and thus a reset with the original drives may help. Once you put the original drives back in without another reset (wiping your data), you'll be running the OS from them again.
Another post seems to suggest turning on flow control for this, but that it should be a problem - could this be connected? Is that advice accurate?
The overrun means that the PC is delivering packets faster than NAS can receive them. Those packets need to be sent again, and that will hurt performance. Enabling ethernet flow control on the PC should resolve that.
You'll find the full ethernet stats in network_settings.log "errors", "frame" and "carrier" indicate hardware issues (NIC port or circuitry in one of the devices or a cable issue).
Also seeing a lot of leafp2p errors but assume that's not connected/ for another day
Oct 15 22:49:56 NETNAS leafp2p[2178]: url https://peernetwork.netgear.com:443//peernetwork/services/LeafNetsConfigWebService Oct 15 22:49:56 NETNAS leafp2p[2178]: SOAP 1.1 fault: SOAP-ENV:Client [no subcode] Oct 15 22:49:56 NETNAS leafp2p[2178]: "Name or service not known" Oct 15 22:49:56 NETNAS leafp2p[2178]: Detail: getaddrinfo failed in tcp_connect()
These are related to ReadyCloud. Since you are directly connecting the NAS to the PC, cloud services won't work. You should probably disable ReadyCloud during this testing (if only to keep the log clutter down).
Re: Ready NAS 102, slow network speed (and saving data on reset)
Something slightly weird.
Having cleared Disk 2 completely; I thought I'd run a quick test and....
Average (W): 64.11 MB/sec Average (R): 80.52 MB/sec
That seems the right ballpark now (?)
I assumed it could be fragmentation of some sort, although
1) I did run defrags before and it didn't improve the speed. 2) Disk 1 hasn't been touched yet but is running faster as well.
So, if you defragment as disk with many shares - do you know whether it will defragment the whole disk properly. Or do multiple shares somehow interfere with that process?
Out of interest, does deframenting the hard drive act upon the OS partitions? Presume not.
With JBOD disks and no snapshots working, should the disk getting full affect speeds? I'm slightly dubious of the disk warnings that talk about degradation of performance. i.e. 10% of a 6Tb drive is still a hell of a lot to play with. But am probably missing something here?
So, if you defragment as disk with many shares - do you know whether it will defragment the whole disk properly. Or do multiple shares somehow interfere with that process?
Out of interest, does deframenting the hard drive act upon the OS partitions? Presume not.
With BTRFS, defrag is done file by file. The NAS defrag function works through the files in the share, and defrags as it finds fragmented files. However, BTRFS sometimes leaves a file fragmented.
The NAS won't defrag the OS partition (which is actually formatted as ext).
There's also the balance function, which can consolidate free space. It's worth running that every couple of months.
