This topic has been marked solved and closed to new posts due to inactivity. We hope you'll join the conversation by posting to an open topic or starting a new one.
Just wanted to reach out to the guru's here as I am stumpped at the moment. I have a ReadyNas 314 connected to a TP-LInk TL-SG108E switch. This switch supports the provision of Link Aggregation Group (LAG) (thats all is says on the box).
Additionally I have an iSCSI connection from a Windows 10 PC using the Microsoft iSCSI initiator.
I would like to add that I have no Apps running on the Readynas 314 and currently at Firmware 6.10.3
Now current config (no bonding configured);
- I have Eth0 plugged into a standard switched port on the SG108E switch (port 3)
- Eth 1 is not connected
- Successful ping to the ReadNas 314
- RN314 is configured for iSCSI and is successfully connected to the Microsoft iSCSI initiator and iSCSI drive is visable
- RN314 is visable within the WIndows 10 Network Neighbourhood
I have figured it out !! (I am pretty excited!). The issue seemed to be with the iSCSI setup on the ReadyNas. I went into the iSCSI settings and within the properties of the group the 'Dedicated Link' was set to 'Any'. Changing this to "bond0:x.x.x.x" completely solved the problem.
In answer to your question acwdc, my SMB settings were as follows; Enable SMB (checked), Legacy Windows Discovery (checked), SMB3 Transport Encryption (Configured Globally - Enabled).
Hope this helps someone else out there. I really appreciate everyones help with this - a big thank you !!
Kind of wondered what would be the difference between ANY and the dedicated bond ID. Anyhow, it world now 🙂
It'd be useful to know what you want to accomplish with bonding. You won't get any performance boost if you only have one client PC running, not matter what teaming mode you use.
Are you connecting to the NAS using it's hostname? If so, perhaps try a test using the IP address instead.
Thank you so much for responding to my query. I do appriciaite your time. The reasons for wanting to implement bondingn on the ReadyNas is to provide some sort of fault tollerance in addition to load sharing.
I have several other computers on the same network all connecting the ReadyNas using their own iSCSI and the network neighbourhood and was attempting to just add a little extra load balancing.
I am connecting to the ReadyNas via the iSCSI initiator via IP address.
No, that won't work. The TL-SG108E only supports static LAGs.
If you are using jumbo frames on the NAS or in the PC, then try turning them off.
think the tplinks use source/dest by mac address as a hash algorithm.
In general, the decision about what packets to put on what link is made by the sender, and is not negotiated. So the switch often does use a different algorithm than the NAS. The hash shouldn't affect connectivity, but it can affect performance.
Good morning everyone, thanks so much for the replies. I think what I may be better off doing is getting another switch that fully supports IEEE 802.3ad.
When enabling Round Robin do I still need to set up a LAG group on the switch or is there any switch configuration that is required at all?
Cheers,
John
Model: RN31400|ReadyNAS 300 Series 4- Bay (Diskless)
I think what I may be better off doing is getting another switch that fully supports IEEE 802.3ad.
I have switches that do support it. The performance gains are usually marginal, unless you have a lot of clients running simultaneously. So I wouldn't get a new switch if I were in your position.
Thanks so much indeed. You know... really I might just discard the whold idea. Seemed like a good idea at the time, but as you mentioned really not going to get much of a performance boost.
It will have to do. Cheers guys and have a great day
Yes that's correct, my mistake, TL-SG108E only supports static LAGs.
Homith, i've got TPlink switches with LAGs configured (static ones... I did not realise this). Just now I've connected a RN312 to one switch and setup a team with XOR/Lv2 only. It works. The only thing I would suggest is to set the config up on the switch and NAS first, then connect the second NIC of the NAS to the switch.
Thanks so much for your reply on this - I do appreciate it. The issue I am confronted with is more to the fact that I can configure LAG on the switch, and set up the ReadyNas to a new bond (XOR/Lv2) and can still receive a ping response from the ReadyNas. I can selectivly disconnect one ethernet port at a time and the ping is constant.
The issue I am facing is that the Windows iSCSI Initiator can no longer reconnect to the target on the ReadyNas and the ReadNas is no longer visable within the Network Neighborhood. Very odd as the ping is constant.
Then, I do not thing you got issues with LAGs. Works on different network layer. I think its got to do with the SMB Master Browser. Have you have Legacy Windows Discovery enabled under the SMB settings of the NAS (i.e. SMB1)? Are all hosts local in your network?
I have figured it out !! (I am pretty excited!). The issue seemed to be with the iSCSI setup on the ReadyNas. I went into the iSCSI settings and within the properties of the group the 'Dedicated Link' was set to 'Any'. Changing this to "bond0:x.x.x.x" completely solved the problem.
In answer to your question acwdc, my SMB settings were as follows; Enable SMB (checked), Legacy Windows Discovery (checked), SMB3 Transport Encryption (Configured Globally - Enabled).
Hope this helps someone else out there. I really appreciate everyones help with this - a big thank you !!
Kind of wondered what would be the difference between ANY and the dedicated bond ID. Anyhow, it world now 🙂
