Jump to content
XPEnology Community

Undetected drives following installation


Salir

Recommended Posts

Hi!

 

First of all, I'd like to thank everyone who is actively involved in and further developing this this project - I love it!

Second of all, I am not sure whether this is the correct thread for my problem, since the loader loader itself actually works like a charm. Please feel free to move this post to a more fitting place, if necessary.

 

As I said, I managed to get a DSM 6.1.4 (15217) - Update 2 for DS3617xs running with the loader 1.02b, however the system does not detect my data drives (anymore, see below).

 

I am running a HPE Proliant Microserver Gen8 with the system disk (SanDisk 120GB SSD) attached to SATA5 (the ODD port on the mainboard, used in RAID0) and 2x SanDisk 250GB SSDs for caching attached to the on-baord RAID controller (HP Dynamic Smart Array B120i) via miniSAS (SATA1 & 2) - those three disks were successfully detected by DSM. Additionally, I am running 4x WD Red 4TB in the built in 4 bay HDD enclosure which is connected to a Startech PEXSAT34SFF (Marvell 88SE92xx chip, SATA6 - 9) in AHCI mode. The controller detects the drives correctly and I was able to access them via a Linux Live-CD, however DSM does not recognize them.

 

For installation, I exactly followed this tutorial and tried several approaches:

* Automatic installation of latest (6.1.4) via download in installation process and automatic update to 6.1.4-Update 2

* Initial 6.1.3 (15152) and manual update (via .pat files) to 6.1.3-Update 8

** Additional manual update (via .pat files) to 6.1.4-Update 2

* Initial 6.1.4 (15217) and manual update (via .pat files) to 6.1.4-Update 2

* Initial 6.1.4 (15217) and automatic update to 6.1.4-Update 2

 

Also I tried with and without having the HDDs initially plugged in during installation and also with the extra RAM disk v. 4.2 for additional drivers.

 

Interestingly enough: In my very first installation approach I was using the automatic installation of the latest DSM (6.1.4) without having the HDDs plugged in (and with the normal RAM disk). After the installation, I hot-plugged my HDDs and they were successfully detected by DSM. I was able to create a volume and successfully copy data on them, but after a reboot of the system they were not detected anymore. Only then I noticed the problem and started debugging this behavior with all the different approaches.

 

I attached my dmesg log which I took from an 6.1.4-Update 2 installation but I saw similar logs with all other installation approaches. There seems to be a problem with responses of the SATA devices. I found out that this was an old Linux kernel problem, hence I updated my BIOS and installed the latest firmware for the server, however, this didn't solve the problem.

 

After spending several hours of searching Google, reading forum posts and bug reports, trying different installations, fiddling with the grub extra arguments and simply (un-)plugging my HDDs I am now at a stage where I cannot think of anything else I could try.

 

dmesg  

 

[    2.123032] ata1: SATA link down (SStatus 0 SControl 300)
[    2.155012] ata12: SATA link down (SStatus 0 SControl 300)
[    2.157017] ata9: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
[    2.157032] ata13: SATA link down (SStatus 0 SControl 300)
[    2.157042] ata8: SATA link down (SStatus 0 SControl 300)
[    2.157055] ata10: SATA link down (SStatus 0 SControl 300)
[    2.157070] ata14: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
[    2.158008] ata11: SATA link down (SStatus 0 SControl 300)
[    2.400884] tsc: Refined TSC clocksource calibration: 2294.786 MHz
[    2.400886] Switching to clocksource tsc
[    2.431868] ata2: SATA link down (SStatus 0 SControl 300)
[    2.738711] ata3: SATA link down (SStatus 0 SControl 300)
[    3.043555] ata4: SATA link down (SStatus 0 SControl 300)
[    3.503317] ata5: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
[    3.507243] ata5.00: ATA-9: SanDisk SDSSDA120G, Z33130RL, max UDMA/133
[    3.507244] ata5.00: 234441648 sectors, multi 1: LBA48 NCQ (depth 31/32), AA
[    3.516821] ata5.00: configured for UDMA/133
[    3.516825] ata5.00: Find SSD disks. [SanDisk SDSSDA120G]
[    3.518552] scsi 4:0:0:0: Direct-Access     SanDisk  SDSSDA120G               Z331 PQ: 0 ANSI: 5
[    3.518769] sd 4:0:0:0: [sde] 234441648 512-byte logical blocks: (120 GB/111 GiB)
[    3.518810] sd 4:0:0:0: [sde] Write Protect is off
[    3.518812] sd 4:0:0:0: [sde] Mode Sense: 00 3a 00 00
[    3.518825] sd 4:0:0:0: [sde] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[    3.519437]  sde: sde1 sde2
[    3.521576] sd 4:0:0:0: [sde] Attached SCSI disk
[    3.824155] ata6: SATA link down (SStatus 0 SControl 300)
[    7.154443] ata14.00: qc timeout (cmd 0xa1)
[    7.154447] ata14.00: failed to IDENTIFY (I/O error, err_mask=0x4)
[    7.154455] ata9.00: qc timeout (cmd 0xec)
[    7.154459] ata9.00: failed to IDENTIFY (I/O error, err_mask=0x4)
[    7.202412] ata7: link is slow to respond, please be patient (ready=0)
[    7.459286] ata9: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
[    7.459298] ata14: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
[   11.892006] ata7: COMRESET failed (errno=-16)
[   11.892102] ata7: COMRESET fail, set COMRESET fail flag
[   17.243260] ata7: link is slow to respond, please be patient (ready=0)
[   17.454151] ata14.00: qc timeout (cmd 0xa1)
[   17.454155] ata14.00: failed to IDENTIFY (I/O error, err_mask=0x4)
[   17.454156] ata14: limiting SATA link speed to 1.5 Gbps
[   17.454162] ata9.00: qc timeout (cmd 0xec)
[   17.454165] ata9.00: failed to IDENTIFY (I/O error, err_mask=0x4)
[   17.454167] ata9: limiting SATA link speed to 3.0 Gbps
[   17.758999] ata14: SATA link up 1.5 Gbps (SStatus 113 SControl 310)
[   17.759009] ata9: SATA link up 6.0 Gbps (SStatus 133 SControl 320)
[   21.932852] ata7: COMRESET failed (errno=-16)
[   21.932948] ata7: COMRESET fail, set COMRESET fail flag
[   27.284105] ata7: link is slow to respond, please be patient (ready=0)
[   47.743607] ata14.00: qc timeout (cmd 0xa1)
[   47.743611] ata14.00: failed to IDENTIFY (I/O error, err_mask=0x4)
[   47.743617] ata9.00: qc timeout (cmd 0xec)
[   47.743620] ata9.00: failed to IDENTIFY (I/O error, err_mask=0x4)
[   48.048451] ata9: SATA link up 6.0 Gbps (SStatus 133 SControl 320)
[   48.048462] ata14: SATA link up 1.5 Gbps (SStatus 113 SControl 310)
[   56.950876] ata7: COMRESET failed (errno=-16)
[   56.950973] ata7: COMRESET fail, set COMRESET fail flag
[   56.951087] ata7: limiting SATA link speed to 3.0 Gbps
[   61.996286] ata7: COMRESET failed (errno=-16)
[   61.996381] ata7: COMRESET fail, set COMRESET fail flag
[   61.996497] ata7: reset failed, giving up
[   61.996585] ata7: get error flags 0x2
[   61.996668] ata7: do detect tries 1
[   61.996747] ata7: device plugged sstatus 0x133
[   61.996752] ata7: exception Emask 0x10 SAct 0x0 SErr 0x0 action 0xf t1
[   61.996896] ata7: irq_stat 0x00000000, PHY RDY changed
[   61.997012] ata7: hard resetting link
[   67.765328] ata7: link is slow to respond, please be patient (ready=0)
[   71.996153] ata7: COMRESET failed (errno=-16)
[   71.996249] ata7: COMRESET fail, set COMRESET fail flag
[   71.996363] ata7: hard resetting link
[   77.765194] ata7: link is slow to respond, please be patient (ready=0)
[   81.996019] ata7: COMRESET failed (errno=-16)
[   81.996117] ata7: COMRESET fail, set COMRESET fail flag
[   81.996234] ata7: hard resetting link
[   87.765059] ata7: link is slow to respond, please be patient (ready=0)
[  117.024039] ata7: COMRESET failed (errno=-16)
[  117.024135] ata7: COMRESET fail, set COMRESET fail flag
[  117.024249] ata7: limiting SATA link speed to 3.0 Gbps
[  117.024249] ata7: hard resetting link
[  122.028470] ata7: COMRESET failed (errno=-16)
[  122.028565] ata7: COMRESET fail, set COMRESET fail flag
[  122.028679] ata7: reset failed, giving up
[  122.028767] ata7: get error flags 0x2
[  122.028850] ata7: do deep tries 1
[  122.028926] ata7: not support deep sleep, do one more detect try
[  122.029060] ata7: device plugged sstatus 0x133
[  122.029064] ata7: exception Emask 0x10 SAct 0x0 SErr 0x0 action 0xf t1
[  122.029208] ata7: irq_stat 0x00000000, PHY RDY changed
[  122.029324] ata7: hard resetting link
[  127.797509] ata7: link is slow to respond, please be patient (ready=0)
[  132.028336] ata7: COMRESET failed (errno=-16)
[  132.028431] ata7: COMRESET fail, set COMRESET fail flag
[  132.028546] ata7: hard resetting link
[  137.797375] ata7: link is slow to respond, please be patient (ready=0)
[  142.028204] ata7: COMRESET failed (errno=-16)
[  142.028299] ata7: COMRESET fail, set COMRESET fail flag
[  142.028412] ata7: hard resetting link
[  147.797242] ata7: link is slow to respond, please be patient (ready=0)
[  177.056223] ata7: COMRESET failed (errno=-16)
[  177.056318] ata7: COMRESET fail, set COMRESET fail flag
[  177.056433] ata7: limiting SATA link speed to 3.0 Gbps
[  177.056434] ata7: hard resetting link
[  182.060654] ata7: COMRESET failed (errno=-16)
[  182.060749] ata7: COMRESET fail, set COMRESET fail flag
[  182.060863] ata7: reset failed, giving up
[  182.060951] ata7: get error flags 0x2
[  182.061035] ata7: ==== port retry failed ====
[  182.061131] ata7: EH complete
[  182.061180] registered taskstats version 1
[  182.061524] rtc_cmos 00:06: setting system clock to 2017-12-10 12:43:22 UTC (1512909802)
[  182.061852] Freeing unused kernel memory: 760k freed
[  182.061910] Write protecting the kernel read-only data: 8192k
[  182.061989] Freeing unused kernel memory: 1320k freed
[  182.062626] Freeing unused kernel memory: 496k freed

 

Hide  

 

 

dmesg_15217-u2.log

Link to comment
Share on other sites

Update:

 

I switched the mode of the on-board RAID controller back to AHCI. Without having my caching SSDs connected, the drives were recognized. After rebooting and connecting the caching SSDs again, the drives were not recognized anymore.

 

Since all three SSDs now are located at port 9, 10, 11 in DSM, I thought that the problem could be the 12 drives limitation of DSM. Hence, I altered the synoinfo.conf (and the one in defaults) to 24, 30 or even 45 drives (which worked), but my HDDs still weren't recognized.

 

The dmesg log is still similar even after switching the mode of the RAID controller which leads me to the assumption that the ports which give errors in dmesg do not fit to the actual SATA ports on my machine. Therefore, ata0 to 8, which gives an error, is just not connect. Then, starting from ata9, my SSDs show up.

 

This behavior now confuses me even more, since I think I tried everything I could find during my research.

Edited by Salir
Link to comment
Share on other sites

pci 0000:00:1f.2: [8086:1c04] type 00 class 0x010400
...
ahci 0000:00:1f.2: AHCI 0001.0300 32 slots 6 ports 6 Gbps 0x3f impl RAID mode

 

0000:00:1f.2 claimed by ahci takes 6 ports - Intel chipset aka HP B120i
1c04 is raid mode, should be 1c02 for ahci mode


pci 0000:07:00.0: [1b4b:9230] type 00 class 0x010601
...
ahci 0000:07:00.0: AHCI 0001.0200 32 slots 8 ports 6 Gbps 0xff impl SATA mode

 

the Marvell 88SE9230, seem to be detected as ahci, 8 ports are more then to expect, it has 4 but at least its detected

 

 

ata0-5 intel
ata6-14 marvell

 

 

[    2.157017] ata9: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
[    2.157070] ata14: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
[    3.503317] ata5: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
[    7.202412] ata7: link is slow to respond, please be patient (ready=0)

only corret detected is ata5, sandisk 120 gb (last port of intel controller)
slow links and loads of errors for marvell ports
if it works with a live linux then it might be some driver problem and thats cant be solved with dsm as we stuck with the kernel and ahci driver
in that case i's sugest trying to install open media vault and see how it does (all disks usable? stable?)
as there are 6 ports from the onboard controller you might use them and get rid of the marvell controller and the ssd's to lower the needed port count
 

  • Like 2
Link to comment
Share on other sites

Thank you for you detailed analysis of my dmesg log. I always wondered which ata ports in the log actually correspond to which controller..

 

3 hours ago, IG-88 said:

if it works with a live linux then it might be some driver problem and thats cant be solved with dsm as we stuck with the kernel and ahci driver

 

I can ensure that the drives as well as the controller card work, as I used Gparted live already extensively and also installed Ubuntu server once. In both scenarios the drives were recognized and showed up as expected. I thought of a driver issue as well, that is why I also tried the extra RAM disk with the additional drivers. Unfortunately, this didn't solve the problem.

Interestingly enough, I was able to get the drives show up in DSM twice already (even with the default ram disk). Therefore, I'd assume that it isn't a driver issue..

 

3 hours ago, IG-88 said:

in that case i's sugest trying to install open media vault and see how it does (all disks usable? stable?)

 

I actually thought about using OMV because of its standard Linux below the hood. Then I remembered the ease of use of DSM on my Synology Diskstation, hence I wanted to install Xpenology. OMV seems like a good alternative, nevertheless I'd like to get this running :D

 

3 hours ago, IG-88 said:

as there are 6 ports from the onboard controller you might use them and get rid of the marvell controller and the ssd's to lower the needed port count

 

I wanted to connect the data drives to a dedicated controller card to make the hot-pluggable and make use of the SATA III interface my drives have. Furthermore, although caching SSDs are certainly overkill, I wanted to make use of this feature which is also integrated in DSM. At last, I also plan to add 2x 1TB 2,5 HDDs for some VMs and other projects which are unrelated from my NAS use.

Edited by Salir
Link to comment
Share on other sites

well it's kind of odd, i was reading some stuff about 9230 and remembered that i'm using a asrock C2550D4I board and that one has 4 port marvell e9230, thing is running fine with dsm 6.1 so it cant be a driver problem

if we rule out power and cable it might be a combination of controller bios and driver?

 

i found something about adding "iommu=soft" to grub config, can't hurt to try and its easy to do/reverse

just edit the grub.cfg on the first partition of the usb flash drive

set extra_args_3615=''

->

set extra_args_3615='iommu=soft'

 

in the end the easiest way will be to look what 4port sata card worked for other people owning a gen8 and use this

as long as we don't use a newer kernel i dint think there is a solution when it depends on the driver, as i said, the ahci driver is part of the dsm kernel and the way dsm is "hacked" we can't exchange the kernel (at least so one did that so far)

 

also overthink whats features you need, omv has plugins too and as its a full linux some stuff is much easier to achieve then with the dsm appliance, dsm/xpenology ,is limited in some areas and will cost a lot of time to make it work in a way you need it, even applying updates can be a problem, if it comes to invested time i'm not so sure that dsm will save you anything, you will just have to invest time in other areas then you would with omv

 

  • Like 2
Link to comment
Share on other sites

  • 2 weeks later...
On 12/16/2017 at 5:11 PM, IG-88 said:

i found something about adding "iommu=soft" to grub config, can't hurt to try and its easy to do/reverse

just edit the grub.cfg on the first partition of the usb flash drive

set extra_args_3615=''

->

set extra_args_3615='iommu=soft'

 

This seemed to fix the issue ... Thanks @IG-88!


I used a clean 3617xs grub.cfg and only added this extra parameter. Then I installed the system by using the latest version of DSM which is directly downloaded from the website. The drives are recognized correctly independent of the slot I put them in (currently I am only using 2 of the 4 slots) even with hot-plugging and also after rebooting.

 

I planned to play with the SataPortMap value in the grub config file (e.g. putting 54 there: 5 drives on the first controller and 4 on the second) but right now the disk number only goes up until 9 when putting the drives in different slots. Though, I need drives for all 4 slots to actually be sure that it will work.

 

On 12/16/2017 at 5:11 PM, IG-88 said:

also overthink whats features you need, omv has plugins too and as its a full linux some stuff is much easier to achieve then with the dsm appliance, dsm/xpenology ,is limited in some areas and will cost a lot of time to make it work in a way you need it, even applying updates can be a problem, if it comes to invested time i'm not so sure that dsm will save you anything, you will just have to invest time in other areas then you would with omv

 

Now that I got Xpenology to run, I want to use it as the main OS on my server, since I made very good experience with DSM on my Diskstation so far. However, I might change to a more generic Linux based system like OMV in the future, when there will be more use cases / application needs for the server. Right now its main purpose is NAS.

Edited by Salir
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...