limetech

Thank you for the report and blog post. It's going to take a while for me to fully digest what you're doing. There's nothing really 'magic' about the Unraid SMB implementation - we use pretty much the latest releases of Samba. You can tweak yourself via Network/SMB/SMB extra configuration.

Not sure the exact issue but something is wrong with your SSL cert. This is not included in diags (obviously) but should be on the flash in config directory: config/ssl/certs named either certificate_bundle.pem or PLEXZONE_unraid_bundle.pem. I suggest you delete any pem files (or rename them to different extension) you find there and then edit the file config/ident.cfg. Find the line that say: USE_SSL="auto" change to: USE_SSL="no" Now your server webGUI should answer up at PLEXZONE.local or use it's IP address. Next you can go to Settings/Management Access and re-enable https (set back to Auto) and let it download a new LetsEncrypt cert.

The problem is that nginx (the http server) is crashing: Jun 10 18:18:53 PLEXZONE emhttpd: shcmd (26): /etc/rc.d/rc.nginx start Jun 10 18:18:53 PLEXZONE root: unable to load certificate Jun 10 18:18:53 PLEXZONE root: 22497627080512:error:0909006C:PEM routines:get_name:no start line:crypto/pem/pem_lib.c:745:Expecting: TRUSTED CERTIFICATE Jun 10 18:18:53 PLEXZONE root: Starting Nginx server daemon... Jun 10 18:18:53 PLEXZONE root: nginx: [emerg] PEM_read_bio_X509_AUX("/etc/ssl/certs/unraid_bundle.pem") failed (SSL: error:0909006C:PEM routines:get_name:no start line:Expecting: TRUSTED CERTIFICATE) Jun 10 18:18:53 PLEXZONE emhttpd: shcmd (26): exit status: 1 Before I offer some suggestions, two things: You mention "creating a new USB boot" - please use your original flash device, without any changes. Next from console/telnet type the command "diagnostics" - you will see that it creates a file on your flash in the 'logs' directory. Please post that file. Please do not call out forum member names just to try and gain their attention.

Right, for NFS the security modes map to entries in /etc/exports as follows (example share name is Test): For Public: "/mnt/user/Test" -async,no_subtree_check,fsid=100 *(sec=sys,rw,insecure,anongid=100,anonuid=99,all_squash) For Secure: "/mnt/user/Test" -async,no_subtree_check,fsid=100 *(sec=sys,ro,insecure,anongid=100,anonuid=99,all_squash) If Private has been selected (and click Apply) a new input field labeled Rule will appear. Whatever you type will replace the third field in the exports line: "/mnt/user/Test" -async,no_subtree_check,fsid=100 <what you type goes here> Hence, for NFS, Public means read/write for anyone; Secure means read-only for everyone; Private gives you control to set whatever you want. The 'fsid' value is established when the share is first created and will retain that value for as long as the share exists.

There have been 5 kernel patch releases since 6.7.1-rc1 and there don't appear to be any problems with Zombieland mitigation. We updated a few security-related patches, but I want to give this a few days of wider testing before publishing to stable. Please post here in this topic, any issues you run across which are not present in 6.7.0 also - that is, issue that you think can be directly attributed to the microcode and/or kernel changes. Version 6.7.1-rc2 2019-06-08 Base distro: curl: version 7.65.0 (CVE-2019-5435, CVE-2019-5436) docker: version 18.09.6 kernel-firmware: version 20190514_711d329 mozilla-firefox: version 66.0.5 samba: version 4.9.8 (CVE-2018-16860) Linux kernel: version: 4.19.48 (CVE-2019-11833) Management: shfs: support FUSE use_ino option Dashboard: added draggable fields in table Dashboard: added custom case image selection Dashboard: enhanced sorting Docker + VM: enhanced sorting Docker: disable button "Update All" instead of hiding it when no updates are available Fix OS update banner overhanging in Auzre / Gray themes Don't allow plugin updates to same version misc style corrections

Waiting for reply from IPS.

On Settings/Global Share Settings there is "Tunable (enable Direct IO)" - default is set to Auto, which means No, but if you set this to Yes it might lead to problems with sqlite. For those experiencing this issue, how do you have this set? Just want to rule out this is the issue.

Can you provide link to that discussion?

Pretty sure unfriendliness was not intended. Would seem so. Many many changes in all kinds of subsystems between 6.6 and 6.7, going to take time to isolate. If you read the entire thread, some are reporting this does not completely solve the problem.

We have been looking into this for the past few days. If you have a way to make this happen on demand, please post how you do it.

FUSE is a Linux subsystem which means "Filesystems in User Space". The Unraid 'shfs' (share file system) component is what implements the union and is built using FUSE. That said, any union file system is not going to be fully POSIX compatible. It would be nice to know what is this new incompatibility, this way we can fix it.

This has been cleared up.

From Redhat description: In general, VM's in general, and PCI-Passthrough to VM's in particular, are highly hardware dependent. If h/w is modern and chosen correctly, this works great, but if bios not up to date, or h/w old or misconfigured, you may end up with far less hair than you started with getting it to work.

Yes

Changed Status to Retest