cooledspirit Posted January 8, 2018 Share #1 Posted January 8, 2018 Hello, running DSM 6.1 (jun's loader) on Asus A88X-Plus/USB 3.1 with AMD A series A10-7860K CPU, 8 Gb RAM and expanded from 8 disks to 10, system is configured as RAID 6. As the motherboard natively only supports 8 disks I added a PCI-E sata controller Highpoint Rocket 640L. I added the 2 disks to the external sata controller. The system appears to have succesfully switched from 8 to 10 disks, but never fully worked,as the system partition appears to be faulty of the 2 added disks. Re-installing DSM did not work either, everything ok until I get a timeout when writing the config files (using network cable, not wireless). The 2 new disks are kicked out for some reason and my system falls down from 10 disk to 8 (RAID 6, so this is critical). After a reboot they are visible again, I can add them and parity check runs but ultimately fails (data ok/system partition fails). Yet the drives don't contain any errors, from extended disk checks from both synology as well as other disk checking software. At this moment the x-th parity check is running after adding the disks again but I fear that this will be yet a futile attempt. What could be wrong ? Is the external sata controller not supported ? Other ideas ? Any tricks I can do to "force" a correct and full system partition re-installation ? Many thanks ! Kind regards, Cooled Spirit. Quote Link to comment Share on other sites More sharing options...
sbv3000 Posted January 8, 2018 Share #2 Posted January 8, 2018 If the controller was not supported (ie no drivers/modules loading) then the disks would not show up so I think thats ok. Some things to check; BIOS settings on the mobo to see if the raid card is sharing resources with other devices, maybe disable unnecessary devices, audio/parallel etc The raid settings on the rocketraid - make sure set to jbod Try moving the drives to the 'other' sata channels and see how DSM reports them as slots - make sure that the drives are on the 'lowest' 2 channels Try the 'sataportmap' setting in the grub config - 84? Check a dmesg output and look for the raid build logs - see if there are timeouts on the drives on the rocket raid assembling in the array Disable drive hibernation in DSM as a test. Check the power supply can support 10 drives + rest of system (startup current as well as operating) Quote Link to comment Share on other sites More sharing options...
cooledspirit Posted January 8, 2018 Author Share #3 Posted January 8, 2018 3 hours ago, sbv3000 said: If the controller was not supported (ie no drivers/modules loading) then the disks would not show up so I think thats ok. Some things to check; BIOS settings on the mobo to see if the raid card is sharing resources with other devices, maybe disable unnecessary devices, audio/parallel etc ==> will do after current attempt The raid settings on the rocketraid - make sure set to jbod ==> will do after current attempt Try moving the drives to the 'other' sata channels and see how DSM reports them as slots - make sure that the drives are on the 'lowest' 2 channels ==> If I move the disks to lowest 2 channels, would it be ok to launch jun's loader for "re-install" and see how that goes (in order to get the system partition correctly installed) ? Or is that a bad idea ? Try the 'sataportmap' setting in the grub config - 84? ==> is there any harm in doing so ? I'll put the value 84 inthere. Check a dmesg output and look for the raid build logs - see if there are timeouts on the drives on the rocket raid assembling in the array ==> I'll send a log after the next reboot. PS: I see mentions of UDMA/133, not exactly the speed I had in mind. I see it kicking out the "non fresh" drives. Disable drive hibernation in DSM as a test. ==> done ! Check the power supply can support 10 drives + rest of system (startup current as well as operating) Thank you very much for the suggestions. Currently another attempt is running, so I have to wait a few more hours before I can check what you suggested. Quote Link to comment Share on other sites More sharing options...
cooledspirit Posted January 9, 2018 Author Share #4 Posted January 9, 2018 Please find the log (dmesg) file in attachment. Please also take a look at my questions in my prior reply regarding running re-install on the disks in the lowest 2 channels ? dmesg-log.txt I disabled the serial port (the USB with JUN's boot loader complained that 3F8 was disabled afterwards) I disabled "powernow" from AMD as well Perhaps running an fsck might help ? But how ? If I run syno_poweroff_task -d, my session gets closed. Quote Link to comment Share on other sites More sharing options...
cooledspirit Posted January 9, 2018 Author Share #5 Posted January 9, 2018 In attachment the dmesg results when booting with only 2 disks in the lowest SATA slots. dmesg - only the 2 disks.txt Quote Link to comment Share on other sites More sharing options...
cooledspirit Posted January 9, 2018 Author Share #6 Posted January 9, 2018 Is "No Caching mode page found" relevant ? Quote Link to comment Share on other sites More sharing options...
cooledspirit Posted January 11, 2018 Author Share #7 Posted January 11, 2018 So this time I have a log that indeed mentions timeouts (see attachment). Can somebody interpret and indicate what might be going on here ? Many thanks in advance ! Cooled Spirit dmesg.log Quote Link to comment Share on other sites More sharing options...
cooledspirit Posted January 12, 2018 Author Share #8 Posted January 12, 2018 This is what happens... last adding of the 2 disks worked (consistency check passed successfully) on 6.1.4 update 1, however as soon as I dare to install 6.1.4 update 5, the system partition appears to fail and the disks are rejected. Any ideas ? Quote Link to comment Share on other sites More sharing options...
cooledspirit Posted January 12, 2018 Author Share #9 Posted January 12, 2018 This is what happens... last adding of the 2 disks worked (consistency check passed successfully) on 6.1.4 update 1, however as soon as I dare to install 6.1.4 update 5, the system partition appears to fail and the disks are rejected. That or simply rebooting is a problem.... Any ideas ? Quote Link to comment Share on other sites More sharing options...
sbv3000 Posted January 12, 2018 Share #10 Posted January 12, 2018 Check the rocketraid bios to see if you manually 'slow down' the sata speed also, disable write cache in DSM storage manager maybe try a cheap non-raid add on controller if you have one lying around Quote Link to comment Share on other sites More sharing options...
cooledspirit Posted January 15, 2018 Author Share #11 Posted January 15, 2018 I disabled write cache again (as re-installing DSM overwrote the setting, I forgot to disable it after reinstalling). I can't find any way in the Rocketraid bios to modify what you indicated. I ordered the following non-RAID sata controller (Syba sd-pex40099 4 Port SATA III PCI-EXPRESS 2.0 x) in the hope this will solve everything. By now, I assume it has something to do with the controller or its driver. I read on another place in this forum that the controller above appears to work. What I notice is: parity check with DSM 6.1, update 1 completes succesfully. The same with update 5 results in failed parity check. Why ? No idea.... DMESG in attachment, maybe it reveals something to the experts. From what I understand, it seems that at some point the driver or hardware freezes up. But since it works for update 1 and not for update 5, something must be different. Can you take a look to the log please ? So for now, I reverted back to update 1 and go for yet another 48 hour parity check, by the time that parity check is done, I will hopefully have the new SATA controller in my possession. dmesg3.log Quote Link to comment Share on other sites More sharing options...
cooledspirit Posted January 16, 2018 Author Share #12 Posted January 16, 2018 I replaced the RocketRaid sata controller can report that all seems to be working fine now. I managed to upgrade to update 5 without any problem. Would it be ok to enable "write cache" again ? Or is that a bad idea ? Quote Link to comment Share on other sites More sharing options...
sbv3000 Posted January 17, 2018 Share #13 Posted January 17, 2018 18 hours ago, cooledspirit said: I replaced the RocketRaid sata controller can report that all seems to be working fine now. I managed to upgrade to update 5 without any problem. Would it be ok to enable "write cache" again ? Or is that a bad idea ? Should be OK to enable again, DSM will probably warn you if it finds a drive where the write caching is problematic (does with some older toshiba drives I have ) Quote Link to comment Share on other sites More sharing options...
IG-88 Posted January 18, 2018 Share #14 Posted January 18, 2018 On 1/16/2018 at 10:36 PM, cooledspirit said: I replaced the RocketRaid sata controller can report that all seems to be working fine now if you want to use the rocketraid 640L you can flash a non raid bios to it http://www.highpoint-tech.com/USA_new/series_rr600-download.htm ans as there is source availible it might be possible to build a driver and use it with dsm, that way the controller might work after all Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.