January 18, 20215 yr Author try again after rebooting. If it still doesn't work, post another set of diags
January 18, 20215 yr 2 minutes ago, Squid said: try again after rebooting. If it still doesn't work, post another set of diags Weird, it works now. Swear I've already done a reboot before. Here my diagnostics if it helps someone track down why it temp broken like that. diagnostics-20210118-1454.zip
January 20, 20215 yr @Squid What do you think about implementing this in common errors plugin? https://forums.unraid.net/topic/46802-faq-for-unraid-v6/page/2/?tab=comments#comment-700582
January 29, 20215 yr Hi all, just upgraded to 6.9.0 rc2 two days ago. I am now getting this error, and I have no idea how to fix it. Anyone have the steps that I can take to fix? Legacy PCI Stubbing found: vfio-pci.ids or xen-pciback.hide found within syslinux.cfg. For best results on Unraid 6.9+, it is recommended to remove those methods of isolating devices for use within a VM and instead utilize the options within Tools - System Devices.
January 29, 20215 yr Author Remove the entries you made within syslinux.cfg to isolate your devices for VM passthrough, reboot (but first disable the VM service) and then go to Tools - System Devices and check off the devices you want isolated.
February 1, 20215 yr Not sure if there is a better place to post this but I wanted to report a minor typo in the plugin. For the warning about the parity check not being scheduled, the "Suggested Fix" starts with, "It is highliy recommended to schedule parity checks . . . " The exact warning text is "Scheduled Parity Checks are not enabled". Thanks for your work.
February 3, 20215 yr Unsure if this has been asked before, as it's not coming up on any search I'm using: I see that I can exclude folders from the scheduled scans, but can I exclude from the extended test? I have a "backup" share that my Windows machines point to, and there are many errors within those backups that Windows doesn't care about, but unRAID does: leading/trailing spaces, etc. Those items are beyond my control to correct, as they are the result of other apps' organization, or even Windows'. I have image backups and file history backups here, and they make "run extended tests" take FOREVER and generate numerous lines in the results that are completely irrelevant. I'd really like to exclude this Backup share from the extended tests. Is there a way to do this? TIA Erik Edited February 3, 20215 yr by ainuke edited mistype
February 13, 20215 yr Quote SSD's are not currently supported within the array, and their background garbage collection *may* impact your ability to rebuild a disk @Squid With "impact the ability" you mean the rebuild speed can be slow? It sounds like it could fail completely.
February 13, 20215 yr Author It means that due to background garbage collection etc the data actually on the SSD may not be what is actually expected, and when rebuilding another drive, corruption may occur. Some drives will work no problems if they support certain things AFAIK.
February 13, 20215 yr 1 hour ago, Squid said: background garbage collection etc the data actually on the SSD may not be what is actually expected We had this discussion here and you can see it's relevant for TRIM and not GC. GC only combines pages of multiple blocks. The data itself stays the same as the LBA is redirected to a page in a new block and the old blocks are marked as unused. Without TRIM the pages of deleted files will stay intact and will be moved by the GC, too. Of course this is inefficient and will slow down the SSD over the time, but it does not influence data consistency. Conclusion: The warning should refer to TRIM and not GC. Maybe something like: Quote SSD's aren't fully supported by now. Do not TRIM your SSD before checking compatibility.
February 14, 20215 yr Author Well, until such time as either @limetech says its ok for SSDs to be in the array under all circumstances or @JorgeB says similar, the test and warning stays, as FCP isn't designed to be definitive, but rather what trips people up.
February 14, 20215 yr I did not say, that the warning should be removed. Only the text should be changed as the current status is "works as long you don't trim or you have a trim compatible SSD". Which, by the way, is something FCP could check through hdparm.
February 14, 20215 yr 18 hours ago, mgutt said: Do not TRIM your SSD before checking compatibility. You can't trim an SSD assigned to the array, trim is disable, on purpose I guess. As for the warning it's better to play on the safe side, but in my experience most SSDs should not cause any issues, the exception from the many I tested is the now discontinued Kingston V300, which caused a couple of sync errors after a power cycle, I would just recommend frequent parity checks, at least in the beginning, if no sync errors are found you're fine.
February 14, 20215 yr On 8/17/2019 at 8:49 AM, JorgeB said: Kingston V300 Maybe a general problem with this model? Someone complained a huge failure rate: Quote We rolled out about 80 PC's with 2x Kingston V300 60GB SSD's in RAID-1. On these installs we see a 10-15% failure rate over a period of max. 3 months, which is off course unacceptable. This is on a PC platform with an on-board Intel ICH7R SATA RAID controller And as a short reminder: Kingston was caught for bait-and-switch with the V300 and freezes caused by a buggy A400 firmware. As you said, it happened only if the server was shutdown, this sounds like defective flash cells not being able to hold their state and nothing related to GC. Are there any other reports? For me it sounds like one buggy SSD model is generalizing a total SSD incompatibility.
February 16, 20215 yr On 1/29/2021 at 8:08 AM, Squid said: Remove the entries you made within syslinux.cfg to isolate your devices for VM passthrough, reboot (but first disable the VM service) and then go to Tools - System Devices and check off the devices you want isolated. Thanks Squid... after sitting on the error for a bit, I gained courage to delete the lines out of /flash/syslinux/syslinux.cfg and it now looks like this, but I am still getting the error in fix common problems... sorry to be a pain, but this looks like the default config. Thoughts? default menu.c32 menu title Lime Technology, Inc. prompt 0 timeout 50 label Unraid OS menu default kernel /bzimage append initrd=/bzroot label Unraid OS GUI Mode kernel /bzimage append initrd=/bzroot,/bzroot-gui label Unraid OS Safe Mode (no plugins, no GUI) kernel /bzimage append initrd=/bzroot unraidsafemode label Unraid OS GUI Safe Mode (no plugins) kernel /bzimage append initrd=/bzroot,/bzroot-gui unraidsafemode label Memtest86+ kernel /memtest
February 17, 20215 yr Author 6 hours ago, ZosoPage1963 said: yep... should I do an extended test? Actually, you'd have to reboot to fix that up.
February 18, 20215 yr interesting, I will reboot again but just to let you know, I live in Houston and recently, the unraid server has been rebooted multiple times due to the power outages here... UPDATE: I have rebooted multiple times due to power outages in my area and the error is still showing. For now, since my syslinux.cfg is exactly like the default, I am ignoring it, but would like to know where else this could be pulling from. I must be missing it somewhere... Edited February 22, 20215 yr by ZosoPage1963 adding reboot info
March 8, 20215 yr Thanks for this plugin, very helpful. My only remaining warning is to install the Dynamix Trim plugin, but according to this thread: I don't need to do trim if it's btrfs and in a pool. Is that correct? If so, could the plugin be tweaked to not give the warning in that case? Thx!
March 8, 20215 yr Getting this error recently Quote Mar 8 11:52:56 unraid root: Fix Common Problems: Error: Multiple NICs on the same IPv4 network My server has 2 nics (well 3 if you count the IPMI nic) and they're both plugged into the same switch. Is this not the way most people have it set up? Is there something wrong with doing it this way?
March 8, 20215 yr 3 hours ago, Dovy6 said: Getting this error recently My server has 2 nics (well 3 if you count the IPMI nic) and they're both plugged into the same switch. Is this not the way most people have it set up? Is there something wrong with doing it this way? See my response here
March 9, 20215 yr Hi @Squid I see Carl VT has posted about this also, but can you look to make Mar 9 07:42:09 Tower root: Fix Common Problems: Warning: Dynamix SSD Trim Plugin Not installed Conditional on pool type is BTRFS if OS >= 6.9?, still needed for xfs.
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.