[[v6.9-b1] pool convert options are based on slots, not actual devices]

This was a deliberate choice. It will fail indeed when not all slots are populated.

 

Let me rethink this ...

 

(thanks for bringing this up)

[[6.8.3]switch view button bug]

"problem" is a big word for this issue, but solved anyway.

[[6.8.3]switch view button bug]

I can reproduce this issue and made a correction for the next release.

Thanks

[[6.8.0] <wsdd high cpu>]

You can always disable this service.

[[6.8.3] SHFS process consuming all memory over time]

I don't know what causes your issue.

 

My main server is up and running for more than 32 days and memory consumption stays steady all the time.

 

[[6.8-RC1] kernel: tun: unexpected GSO]

For the upcoming version the default driver is changed to "virtio-net".

 

This means newly created VMs will use the revised driver, but existing VMs need manual adjustment.

[[6.8.3] CPU Isolation Banner]

8 minutes ago, jsebright said:

Could it be the orders were different (2,4,8,10 vs 2,8,4,10 ?)

That's indeed the issue. The automatic check generates a list in sequence for comparison.

When you clear and re-appy then the settings themselves will also be in the correct sequence.

[[6.8.3] CPU Isolation Banner]

According to your diagnostics, the system has been started with

  isolcpus=2,4,8,10
 

A check is done if in syslinux.cfg the same setting is present under the "default" section, if not the banner is displayed.

 

Do you manually change the boot menu selection when starting the system?

[[6.8.3] Network traffic hitting incorrect L3 interface.]

Additional default gateways are usually not needed, but if they exist and with the correct metrics set, they will be used only when the more preferred choice is not available. I.e. an interface goes down and a second interface exists as backup.

 

When DHCP is used to obtain an IP address, your router will automatically add a default gateway to each interface/VLAN, which results in having multiple default gateways.

In the next release of Unraid, it is possible to set metric=0 for additional interfaces/VLANs, which let the system ignore the default gateway announced by the router.

With this approach you can let your main interface (br0) get a default gateway, while all other interfaces/VLANs just get an IP address by DHCP and avoid all kinds of routing problems.

 

[[6.8.3] CPU Isolation Banner]

46 minutes ago, testdasi said:

If you changed the isolation in flash settings, it will also show that banner if you go to the CPU Pinning page. I think there's some config in the CPU Pinning page that saves the boot isolcpus and then compared to the current settings and if diff then show banner.

Correct, there is a check done to see if configured and actual settings (syslinux.cfg) are the same. If not the banner is shown.

[[6.8.3] Network traffic hitting incorrect L3 interface.]

Setting metrics using DHCP is a dynamic interaction. There is no such thing as first come, first served.

In other words there are no guarantees that the first interface, which comes up, will get the lowest metrics.

 

Also the system itself can not know which priority is needed, it treats all interfaces equal. Only the user can determine what is really desired.

 

Usually, it is not needed to have additional default gateways. E.g. a VLAN interface is a local network without internet access. And in such a case a static entry without default gateway can be configured.

 

The correct way to set priority, is to set the metric for each interface and VLAN specifically in case multiple default gateways are defined (using DHCP).

 

In your case specify a higher metric for each VLAN that you have created.

[Nic down to 100 Mb/s from 6.6.7 to 6.8.3 or 6.9]

Realtek driver support in Linux is hit and miss. For some reason Realtek doesn't seem to give much priority to it.

 

You can either stick to a working version, or hope they have corrected things in a newer version.

 

My general recommendation is to go with an Intel NIC

 

[Logspam = 100% full /var/log, things stop working in WebUI]

This is an issue with your system, not a bug.

Please open a topic under general support.

 

[[6.8.3] docker image huge amount of unnecessary writes on cache]

On 3/21/2020 at 10:36 AM, S1dney said:

I've been using this for a while and have been very happy with it so far.

This is really not the recommended way to solve your issues.

 

rc.docker gets updated with new versions of Unraid and this will certainly break old versions.

 

It is always possible to make a feature request and do a proposal for improvement

(sorry didn't go through all details here, so not sure what the actual changes are)

[Unraid OS version 6.9.0-beta1 available]

24 minutes ago, eagle470 said:

I ran into this issue as well. I originally thought it was BOINC causing the problem, but I had it happen after I ran a parity check and had turned off BOINC and F@H. So I ended up reverting back to 6.8.3 for now, looking forward to a more stable version in the future!

Sorry, but this doesn’t add anything. If you want to test beta versions, you need to be prepared to investigate and put more info to the table.

 

Simply stating “I have an issue” doesn’t give us anything to work on.

 

Most instability issues happen due to underlying hardware issues. It is always important to rule this out first.

[6.9.0 beta1 syslog messages when vm and docker on br0]

29 minutes ago, alturismo said:

ok, thanks for the info, so no workaround to get the VM´s use another custom br ... to bypass this behaviour

The workaround is docker containers and VMs don't share the same netwerk interface (VLAN), but use separated interfaces (VLANs) instead.

 

As said, it is a workaround but not the solution.

[[6.8.2] CPU usage reported by GUI, is way different than True CPU usage]

If you don't want to get this lost, better make a feature request and it will stay visible.

[[6.8.2] VLANS cause errors to appear on Docker Settings tab]

Please retest in safe mode.

 

VLANs is working fine.

 

[[6.8.2] CPU usage reported by GUI, is way different than True CPU usage]

This has been explained here on the forum.

 

The GUI includes I/O wait in the CPU load, which the other tools don't do.

 

Not a bug.

[Terminal Window turns 'white' after 20 seconds]

5 hours ago, zaquarian said:

but saw that ghostery attaches its classes to the same element as the <terminal>

Why not removing "ghostery" then?

[writing share setting sto all other drives only does next drive]

Yes, this is a known bug and will be fixed in the next version.

[6.8.2 System Crashing]

Please open a support topic instead of this bug report.

[Update OS display issue]

Thanks, corrected in next version

[6.8.0 SMB Ver 4.11.3 significant performance decrease when opening files in folders with 1000+ files in them.]

I have these entries in the "Samba extra configuration", see Settings -> SMB -> SMB Extras

[global]
case sensitive = true
default case = lower
preserve case = true
short preserve case = true

 

[[6.8.2] Docker Container not listing all br's]

1 hour ago, civic95man said:

I believe this is my case as well

yep, duplicate assignments

1 hour ago, civic95man said:

Now is this just common knowledge passed from person to person, or is it documented somewhere

While doing the implementation for Unraid, I ran into multiple issues limitations of Docker. I don't know if they have these all documented.

1 hour ago, civic95man said:

So it seems my only solution would be to change the subnet for eth1

If your router allows adding a "secondary address" to its LAN interface, then you are good to go.

Or perhaps your router supports multiple LAN ports, each with their own network assignment.

 

Remember that more complex network setups on Unraid does require network equipment (your switch / router) to support that.