thany

Weird, after enabling the crappy onboard NIC, I can see eth0 appearing in ifconfig, but still no network access at all. So this is getting kind of urgent now... The onboard NIC used to work fine as well before I upgraded to the 10Gb card. Halp

So I just upgraded to 6.9.1, and now there's zero network access. I can't find the eth0 in ifconfig even, and so pinging the server doesn't work, and pinging from the server also doesn't work. I would attach my diagnostics, but I'm not sure how to transfer them over without network access 🙈 I have two NICs: one is a onboard 1Gb thingy, which probably works, but is disabled in the UEFI. The other is a 10Gb NIC add-in card. It's a Buffalo LGY-PCIE-MG, but it worked fine in 6.8.3. For this reason, I'm guessing the upgrade must've done something to mess up the driver maybe? What's the best way to fix this? Or, in the meantime, (and this is kind of important, as my server is totally useless now) what is the recommend downgrade path, if a fix or workaround is not currently available?

Can you explain how to do this? I know what a SRV record is, and I know how to add one. But what should it contain exactly? Should I put "_matrix._tcp.matrix.example.com" in there, given "example.com" is my domain? Or do I put in "_matrix._tcp.example.com"? And what are those "_matrix" and "_tcp" parts? And where do I put port 443? What does DNS have to do with port numbers?

When I try that, I get this: $ register_new_matrix_user -c /data/homeserver.yaml http://0.0.0.0:8008 New user localpart [matrix]: Password: Confirm password: Make admin [no]: yes Traceback (most recent call last): File "/usr/local/lib/python3.7/dist-packages/urllib3/connection.py", line 170, in _new_conn (self._dns_host, self.port), self.timeout, **extra_kw File "/usr/local/lib/python3.7/dist-packages/urllib3/util/connection.py", line 96, in create_connection raise err File "/usr/local/lib/python3.7/dist-packages/urllib3/util/connection.py", line 86, in create_connection sock.connect(sa) ConnectionRefusedError: [Errno 111] Connection refused During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/usr/local/lib/python3.7/dist-packages/urllib3/connectionpool.py", line 706, in urlopen chunked=chunked, File "/usr/local/lib/python3.7/dist-packages/urllib3/connectionpool.py", line 394, in _make_request conn.request(method, url, **httplib_request_kw) File "/usr/local/lib/python3.7/dist-packages/urllib3/connection.py", line 234, in request super(HTTPConnection, self).request(method, url, body=body, headers=headers) File "/usr/lib/python3.7/http/client.py", line 1244, in request self._send_request(method, url, body, headers, encode_chunked) File "/usr/lib/python3.7/http/client.py", line 1290, in _send_request self.endheaders(body, encode_chunked=encode_chunked) File "/usr/lib/python3.7/http/client.py", line 1239, in endheaders self._send_output(message_body, encode_chunked=encode_chunked) File "/usr/lib/python3.7/http/client.py", line 1026, in _send_output self.send(msg) File "/usr/lib/python3.7/http/client.py", line 966, in send self.connect() File "/usr/local/lib/python3.7/dist-packages/urllib3/connection.py", line 200, in connect conn = self._new_conn() File "/usr/local/lib/python3.7/dist-packages/urllib3/connection.py", line 182, in _new_conn self, "Failed to establish a new connection: %s" % e urllib3.exceptions.NewConnectionError: <urllib3.connection.HTTPConnection object at 0x1544ebed7198>: Failed to establish a new connection: [Errno 111] Connection refused During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/usr/local/lib/python3.7/dist-packages/requests/adapters.py", line 449, in send timeout=timeout File "/usr/local/lib/python3.7/dist-packages/urllib3/connectionpool.py", line 756, in urlopen method, url, error=e, _pool=self, _stacktrace=sys.exc_info()[2] File "/usr/local/lib/python3.7/dist-packages/urllib3/util/retry.py", line 573, in increment raise MaxRetryError(_pool, url, error or ResponseError(cause)) urllib3.exceptions.MaxRetryError: HTTPConnectionPool(host='0.0.0.0', port=8008): Max retries exceeded with url: /_matrix/client/r0/admin/register (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x1544ebed7198>: Failed to establish a new connection: [Errno 111] Connection refused')) During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/usr/local/bin/register_new_matrix_user", line 20, in <module> main() File "/usr/local/lib/python3.7/dist-packages/synapse/_scripts/register_new_matrix_user.py", line 221, in main args.user, args.password, args.server_url, secret, admin, args.user_type File "/usr/local/lib/python3.7/dist-packages/synapse/_scripts/register_new_matrix_user.py", line 134, in register_new_user user, password, server_location, shared_secret, bool(admin), user_type File "/usr/local/lib/python3.7/dist-packages/synapse/_scripts/register_new_matrix_user.py", line 43, in request_registration r = requests.get(url, verify=False) File "/usr/local/lib/python3.7/dist-packages/requests/api.py", line 76, in get return request('get', url, params=params, **kwargs) File "/usr/local/lib/python3.7/dist-packages/requests/api.py", line 61, in request return session.request(method=method, url=url, **kwargs) File "/usr/local/lib/python3.7/dist-packages/requests/sessions.py", line 542, in request resp = self.send(prep, **send_kwargs) File "/usr/local/lib/python3.7/dist-packages/requests/sessions.py", line 655, in send r = adapter.send(request, **kwargs) File "/usr/local/lib/python3.7/dist-packages/requests/adapters.py", line 516, in send raise ConnectionError(e, request=request) requests.exceptions.ConnectionError: HTTPConnectionPool(host='0.0.0.0', port=8008): Max retries exceeded with url: /_matrix/client/r0/admin/register (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x1544ebed7198>: Failed to establish a new connection: [Errno 111] Connection refused')) So that's not good, is it? How do I make it not refuse the connection?

This. Synology could find usecases for this, because the usecases are there. What they are, is open to speculation, and not awefully relevant to the question whether we should have this feature in a future version, or not. Also keep in mind that if you specifically DON'T want a file manager built graphically into the WebUI, power to you - you're free to ignore it, and it won't get in your way. Everyone is different, and since unRAID is being used in a variety of scenarios, you can't say something always makes sense, or indeed never makes sense. I think this topic has proven that many people would like this built into the WebUI (and it could even be offered as an official plugin) so it makes sense to, at the very least, get this onto some sort of todo list for the authors, if not already.

Oh dear, you're right 😳 So sorry about that, I should've probably checked this first. I do wonder what `br0` was doing with the IP-address then. I understand that's not a physical NIC, but should `br0` get the IP-address instead of the physical NIC?

Here you are. andy2-diagnostics-20201007-2243.zip

I've set up a fresh new unRAID 6.9.0-beta, latest version from today. I've given it a static IP-address. I've double-checked that its settings are correct. Correct subnet, netmask, gateway, etc. It's all correct. Chosen IP-address is not in use by something else, as far as I can tell. I've checked in the local console using `ifconfig` that indeed, `br0` is given the IP-address I chose. The remaining interfaces, `bond0` and `lo0` have no IP-address and the loopback address, respectively. So it looks to me like it's configured correctly. On the local console, I cannot ping any of my networked devices, including the gateway. It's a simple network, can't really mess it up. I also can't ping from any of my devices to the unRAID server. The network chip should be a Realtek RTL8111H, which is built onto the motherboard. The motherboard is an ASRock B550M Pro4. I don't know how to proceed now. Can anyone give me some pointers on what to do? Or if you need anything in the way of diagnostics, just let me know which logfile to grab and I'll try posting it.

It errors for me as well, telling me links aren't supported. Can you please make a version where MongoDB is integrated into the package? This would certainly avert a lot of problems (see above) and would make it easier to get started. The way I understand Docker, is that a docker should be self-contained. So it should contain everything it needs to run, and only go to the "outside world" for file system storage, and configuration files. If you are unable to do this, could you at least make an easy to follow "getting started" guide? Docker being Docker, and unRAID being unRAID, it should be more or less identical for nearly everyone.

Default settings should always at least do *something*. If default settings give all kinds of errors, or plainly don't work at all, they are wrong. All unRAID installations concerning the docker daemon, are virtually identical. This docker is in the CA repo, so it ought to work. If special instructions are vitally important, they should be more visible right then & there, before/while/after installing. This is not something you should do, of course, assuming you didn't build the CA plugin. Not sure where to report that, so I hope a CA developer is reading along.

But whatever the docker comes with, should at least work. Settings can be adjusted later, but for a bog standard install, it should just work.

OnlyOffice is super slow to install itself. The docker image installed just fine, but when first opening the web UI, of course onlyoffice needs to set itself up. This is taking forever! It's currently using about one of my four cores (on average) and I'm not sure where I can see I/O usage. But it's definitely doing something. How long does it normally take? Given that it's singlethreaded, it shouldn't make very much difference from system to system how long it needs for this process. I mean, is it minutes, 10's of minutes, hours? Just a ballpark estimate would be nice to know.

Just installed UNMS, but when I open the web UI, my browser says "Secure Connection Failed" with code SEC_ERROR_INADEQUATE_KEY_USAGE.

Are those neccesary settings not the default settings then? If not, they should be, right?

No amount of updates to Docker and the plugins will update the core system with new packages that may contain bugfixes and security updates... 22 Beta's, wow! Someone is being a busy bee I just hope the list of "what's changed" will be boiled down to the big changes only. All those updated packages, I'm okay with that. All those tiny fixes, those are all fine. But big new changes might be important, and might be overshadowed by the very many small fixes. But I digress. It's good to see there's still acivity. I was just thinking, since the updates "stopped" since March, the whole corona situation might have had something to do with it. Good thing Mr. Limetech is still going strong

No I think you misunderstand. I didn't add a second card per se. I *replaced* the existing card. There was more than one NIC in the system only once, but the old one was replaced in the end. If there's a single NIC in the system, I shouldn't have to tell it anything, other than setting the IP-address and such.

It's not working. Just did a plain vanilla default installation. So no need to post my config - it's absolutely bog standard. Opened the web UI and... "Unable to connect" It just doesn't work.

And exactly how do I do that when I can't have both cards in the server at the same time? And also, why do I need to tell the OS how to do its internal workings? It should figure it out by itself.

The latest update, 6.8.3, was from 2020-03-05. That's a while ago. Surely, if not unRAID features and bugfixes, the underlying linux packages have had loads of updates in the meantime. Can we please get slightly more frequent updates? Like monthly at least? Those linux packages don't get updated for no reason...

I've got a Basic license currently. Good enough. But I'm thinking of expanding my setup with an 8-bay NAS function, so I should need to upgrade to a Plus license. But I was thinking. What is the actual difference between Basic, Plus, and Pro? Surely, it can't be just upping the arbitrary limit of storage devices, can it? If that, is it an arbitrary limit in the first place, or is inherently more difficult or does it require very different logic, to support more than 6 storage devices? Or does a bigger license grant other advantages that I'm not seeing? It's seems kind of uneven to me that it costs $39 for an upgrade that turns a 6 into a 12. So please let me know how this works.

Well, it works. Kind of. I don't see any intermittent dropouts, I don't see packet loss, and I do see a 10000Mbps link in the dashboard. It took some fidling, but it seems unRAID is not happy about changing NICs in general - unlikely that it has anything to do with this particular card. There's just one tiny little itty bitty thing. I'm not 100% sure, but it seems like I can't get any speeds over 1Gb/s. The only viable device in my network, at least for now, is a NAS on a 4x1Gb link aggregation, and that should give it a 4Gb/s combined link. It and the switch are doing that using 802.3ad LACP, which should bundle the 4 NICs instead of spreading them across multiple clients. The switch uplinks to the unRAID server on a true 10GbE connection, as proven by unRAID's dashboard. It should work, but the speeds I'm getting are disappointing at best for a 10GbE card, or even for a 4Gb link. And mind you, my server has SSD's in them, comfortably able to saturate the SATA interfaces. The NAS has 5 drives, capable of 150MB/s each, in a kind of RAID5 configuration, so they too should comfortably be able to get past those pesky 115MB/s. But the card at least works.

Here's what I did: Added a secondary NIC Started the server Gave the new NIC an IP address I can access Went to that IP-address Switched the two IP addresses around so that the new NIC gets the IP address I want to access the server by. Reboot, just in case, and disconnected the old NIC The result: Absolutely no network access. The startup screen at the local console tells me the IP address the server listens on, which doesn't work. I login to figure out what the heck is going on in `ifconfig` and I discover that my new NIC is eth1. Maybe that's why. So then, I remove the old NIC from the computer, since the new one appears to be working fine anyway, and start her up again. It's still on eth1. And the problem persists. Here's what came next: I changed the `NAME` in `/etc/udev/rules.d/70-persistent-net.rules` to eth0. The original eth0 was gone at this point anyway. And reboot for good measure. Then the server came back to life. Well, almost. I found that the server can't access the internet. In the network setting (now in the web interface) everything appears to be fine. Back at the local console, I CAN ping other computers in the nerwork, and I CAN ping the gateway. I CANNOT ping anything beyond my house though. I suspected it's not picking up the DNS servers or something. So I decided to delete config/network.cfg from the flash. And it came back alive again, but this time on DHCP. It was also able to access the internet properly, but I don't want a DHCP-assigned address. I want a static one. How hard can it be, right? So back to the network settings. Changed it from DHCP to Static, and filled out my desired IP address. All other settings were already filled in correctly. I hit Apply. I wait. And wait. It shouldn't take more than a second or two to change an IP address, should it? Okay, a minute at most, right? Five minutes? Back to the local console. Reboot the darn thing then. That also take ridiculously long. After 10 minutes of trying to shutdown, it's still on whatever phase after unmounting /boot. I'll try the reset button then. Let's see. The server came back to life. It's probably not too happy about that, and it'll probably retaliate a few months from now. However, the IP address is correct, it's using my new NIC, and it can access the internet. Story time's over. Now, why did that have to be so hard?

I've ordered the BUFFALO LGY-PCIE-MG, which uses a Tahuti Networks TN4010 chipset. Not sure if it'll work, but I'll give it a go. If it doesn't work, I'll use it in another computer. If it does work, I can order an additional one I thought I'd mention it here, because the chipset might be similar, and driven by the same driver as the one mentioned in the TS.

Okay, so for most intends and purposes, there is no such thing as an "unRAID driver". I assumed there was

I'll give it a few more days In the mean time, would the best other option to go with the (as I've read) hothead ASUS one, or a more expensive Intel card? The trouble is, in my server case I need a low-profile card, and I'm not 100% sure which cards come with a suitable bracket.