Diverge

It takes a lot less work to actually read the thread

If you don't login WebUI, it will not trigger the unmount action. So everything looks fine. Is there any way to figure out which process, app, is causing it at login? Maybe we can just replace it with previous build.. but I doubt it would be that easy.

I guess that explains why I couldn't find my drives via shell. I tried to look through dmesg for anything regarding unmounting the volumes, drives, ect, but didn't see anything (but I'm kinda noob to this stuff). I guess they moved security into DSM itself, rather than just the kernel. Hopefully someone figures out a work around.

Don't try to update, synology added new security measures, it will unmount your volumes. edit: made warning bigger. If you do update, do it in a test VM

doesn't work for me. That method is only for small updates, not new builds (4458 vs 4482). edit: I think gnoboot just needs to and the new build version to the version file buried in rd.gz, which is buried in the zImage now. Lets hope he finds some time to update it edit2: looks like I was wrong. It's much easier Just need to edit grub menu_me kernel /zImage gnoboot_me=5.0-4482 But looks like they added new secruity meaursures

you can shorten the command to the following: sed -i 's/flashupdateDeb/flashupdateDeb1/' /autoupd@te.info -i flag is an in-place edit. no need to create a temp file and copy over old one.

For my motherboards, both Intel brand, I need to rmmod=ata_piix in grub. Or else DSM will start the array at slot 3 and onward (it seems to reserve the first 2 slots for that driver for some reason). You can try that in a test VM and see if your array starts at the correct slot. If not, then it's probably another driver. edit: I guess it happens for me because I install gnoboot to a IDE vdmk rather than ACHI. https://ata.wiki.kernel.org/index.php/Ata_piix On a side note, does anyone know if it's possible to run gnboot off a .flp (floppy image) rather than virtual hard drive? I tried to get it working, but got boot errors.

303-1E03 1st character = month in hexadecimal, lower case (1=Jan, … , a=Oct, b=Nov, c=Dec) 2-3 = month in decimal, zero padded and starting in 1 (01, 02, 03, …, 11, 12) 4 = dash 5-6 = day of the month in hex (01, 02 .., 0A, .., 1F) 7-8 = greatest common divisor between month and day, zero padded. This is always a number between 01 and 12. I can't help you on your issues, I run my system in esxi.

I have same issue. I dunno how to get rid of it either.

You are right, it writes to /volume1 and might fill up your disk. Just move the core files to another directory such as /tmp or /dev/null, http://justlearningtoblog.wordpress.com/2010/07/23/enable-core-dump-files-in-fedora/ Add this line to your /etc/rc.local. sysctl -w kernel.core_pattern=/tmp/core.%p Thanks for the tip. I set it to null, I don't want unnecessary writes. I removed old files and now none get created

Anyone have this stuff going on at the end of thier dmesg? [ 272.026464] dsmnotify.cgi[18597]: segfault at 0 ip 000000000804ccb8 sp 00000000fffd2970 error 6 [ 272.074134] storagehandler.[18598]: segfault at 0 ip 000000000809e298 sp 00000000ff96e2e0 error 6 [ 280.920198] dsmnotify.cgi[18861]: segfault at 0 ip 000000000804ccb8 sp 00000000fff69c50 error 6 [ 280.935442] dsmnotify.cgi[18862]: segfault at 0 ip 000000000804ccb8 sp 00000000ffe36a10 error 6 [ 281.046709] dsmnotify.cgi[18864]: segfault at 0 ip 000000000804ccb8 sp 00000000ff81b720 error 6 [ 311.439705] dsmnotify.cgi[18940]: segfault at 0 ip 000000000804ccb8 sp 00000000ff9810f0 error 6 [ 341.979578] dsmnotify.cgi[19070]: segfault at 0 ip 000000000804ccb8 sp 00000000ffa52e20 error 6 [ 372.845913] dsmnotify.cgi[19121]: segfault at 0 ip 000000000804ccb8 sp 00000000ffb052b0 error 6 [ 403.800579] dsmnotify.cgi[19235]: segfault at 0 ip 000000000804ccb8 sp 00000000ffb33260 error 6 [ 435.754667] dsmnotify.cgi[19290]: segfault at 0 ip 000000000804ccb8 sp 00000000ffab1590 error 6 [ 467.728926] dsmnotify.cgi[19423]: segfault at 0 ip 000000000804ccb8 sp 00000000ffb5e750 error 6 [ 498.675152] dsmnotify.cgi[19462]: segfault at 0 ip 000000000804ccb8 sp 00000000ff8146c0 error 6 [ 531.624708] dsmnotify.cgi[19595]: segfault at 0 ip 000000000804ccb8 sp 00000000ff9f9d20 error 6 [ 563.602589] dsmnotify.cgi[19643]: segfault at 0 ip 000000000804ccb8 sp 00000000ff921900 error 6 [ 595.548976] dsmnotify.cgi[19767]: segfault at 0 ip 000000000804ccb8 sp 00000000ffc42230 error 6 [ 627.502107] dsmnotify.cgi[19828]: segfault at 0 ip 000000000804ccb8 sp 00000000ff9d0970 error 6 [ 659.214933] dsmnotify.cgi[19966]: segfault at 0 ip 000000000804ccb8 sp 00000000ffa40ff0 error 6 [ 689.989727] dsmnotify.cgi[20050]: segfault at 0 ip 000000000804ccb8 sp 00000000ffc3f590 error 6 [ 720.084700] dsmnotify.cgi[20185]: segfault at 0 ip 000000000804ccb8 sp 00000000ff8eece0 error 6 [ 750.743911] dsmnotify.cgi[20275]: segfault at 0 ip 000000000804ccb8 sp 00000000ff80eb20 error 6 [ 781.235727] dsmnotify.cgi[20414]: segfault at 0 ip 000000000804ccb8 sp 00000000ff87e3a0 error 6 Whatever the issue is, it's an ongoing thing... more entries over time. And it writes to associated files in /volume1/ each time... so it causes lots of writes. I'm on gnoboot 10.4, in esxi 5.5, using LSI 9201-8i in passthrough (VTd) w/ 4x 3TB WD REDs. I've seen it with virtual disks, w/ no HBA or passthrough as well.

For each release I make a new vmdk (I'm using esxi), and just swap them out. On first but it updates whatever new kernel changes there are.

Just a heads up for anyone updated to update_1 on a system open to the internet; there's was a post on official synology forum about possible security bug in this update... I don't know how true it is, but here's the link to the guys complaint: http://forum.synology.com/enu/viewtopic ... 15#p313327 Maybe he did something wrong... I dunno. I use passwords, and I already had SSH enabled on my test VM

I just found a new way to install small updates. On synology's FTP server you can find the small updates packaged as a .pat file, in a critical update folder. download it, then do a manual update. It will get to a prompt that applies the update, that's when you do the sed script modifications sed 's/flashupdateDeb/flashupdateDeb1/' /autoupd@te.info > /autoupd@te.info1 mv /autoupd@te.info1 /autoupd@te.info then click yes, and it will update

I have a problem that I'm hoping someone has experience with, and maybe a solution. I have a VM setup with synology DSM. It has VTd, and has a LSI 9211-8i passed through. I have a 5 hard drive bay, and disks 1-4 are part of the DSM array. The 5th drive is configured as esata via the /etc.defaults/synoinfo.conf file. The problem is if the drive is connected when DSM is booting up, it seems to randomly (?) pick the first 4 disks that get initialized for starting the array. The first time I booted up with all 5 disks in, it booted to a degraded array because it used the wrong disk, the one for backups, as an array disk. I powered down, removed the backup drive, powered on, and the array was still marked as degraded... I guess once it gets marked it can't be reversed without repairing it. Anyway, now I make sure to remove the drive before booting up... Anyone have any solutions to my problem? Or another way to make a specific drive seen only as esata by DSM?

read through the comments on his blog. he posted it there. if people didn't comment so much asking for it, they might see it.

Beta 7 might now have correct driver. I'd try a newer version. Read gnboots blog to find out which versions have which drivers, ect.

I'm seeing some weird stuff with gnoboot 10.4, which I don't think I've seen in past releases. If I boot up a VM with stock grub settings, drives will start off at slot 3 (sdc), since I didn't rmmod=ata_piix in grub yet. I create a test volume (on VM disk), all is good. Then I power down. Edit the grub to remove ata_piix, boot up and my volume is gone, but first disk in slot 1. Create volume, all good. Power down. Change grub back to stock. Power up, disk start at slot 3, and volume still intact - all good. Power down. Change grub again to remove ata_piix, power up. Volume gone again... So it seems if the volume is created, and the disk move up slots, it's okay. But if they move back, then volume goes poof. I'm not sure if the issue is DSM 5 (official) or gnoboot 10.4, but I've never seen a volume disappear after removing ata_piix before. When I say disappear, I mean disk there, but volume data lost in DSM. edit: rebooted VM with volume in MIA state, and on reboot it showed crashed this time. edit2: maybe this is related to the note about mpt2sas. "Disks are not properly initialize when using mpt2sas driver and VT-d during first boot of ESXi host, reboot VM until HBA is detected". I'm not using my HBA in this test VM, but maybe it has similar problem and when the volume was MIA (missing, but not marked as crashed yet), it was because disk wasn't properly initialized

each drive is broken into at least 3 physical partitions per disk. The 1st partition of each drive makes up /dev/md0, which is where DSM gets installed. It's raid 1. The 2nd partition of each drive makes up /dev/md1, which is swap file. It's raid 1. The 3rd partition of each drive makes up /dev/md2, which is data; /volume1. It's raid 5. ... add additional partitions for additional data volumes, expanding arrays by using bigger disks, ect. here's a link that explains linux software raid and LVM https://wiki.archlinux.org/index.php/So ... ID_and_LVM

Did you use the method here for small updates? viewtopic.php?f=2&t=2049 ? I'm not on 4.3, but never had a problem using it with beta 5. The steps leading to this: sed 's/flashupdateDeb/flashupdateDeb1/' /autoupd@te.info > /autoupd@te.info1 mv /autoupd@te.info1 /autoupd@te.info you can put it in a scheduled task, after downloading the update, run the task, then hit update.

that board doesn't have pcie, it's pci (no e).

smartmontools (smartctl) works with RDM drives in terminal of DSM, but DSM uses synodisk (synology tool) to read temps, and it returns -1 when I guess it doesn't have access to the controller.

I will have to wait till Trantor does his version. But thanks to people for their help If you have no monitor, you could always make another copy, but edit the grub menus for the upgrade path. then once upgraded switch to regular version.

If he's posting a kernel patch, he probably built his own

Don't let ungrateful jerks piss you off. Many people just come to the forums to ask for stuff without even reading anything, or trying to learn. They just want stuff handed to them.