September 21, 200817 yr checking temp. at this point, using: smartctl -d ata -A /dev/x and grabbing the 4th field from the "Temperature_Celsius" record. I use the 10th field of the same line for my unmenu.awk script. You could use something like this: smartctl -d ata -A theDisk | grep -i temperature | awk '{ print $10 }' Joe L.
September 21, 200817 yr Author Olympia, are you able to telnet into your unraid box and give me the output from: smartctl -d ata -A /dev/X replacing "X" with one of your disks (sda, sdb, etc.) Also, if you could give the full notification email sent.
September 21, 200817 yr smartctl -d ata -A /dev/X output root@Tower:~# smartctl -d ata -A /dev/sdd smartctl version 5.36 [i486-slackware-linux-gnu] Copyright (C) 2002-6 Bruce Allen Home page is http://smartmontools.sourceforge.net/ === START OF READ SMART DATA SECTION === SMART Attributes Data Structure revision number: 16 Vendor Specific SMART Attributes with Thresholds: ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 1 Raw_Read_Error_Rate 0x000f 100 100 051 Pre-fail Always - 0 3 Spin_Up_Time 0x0007 100 100 015 Pre-fail Always - 7232 4 Start_Stop_Count 0x0032 099 099 000 Old_age Always - 1332 5 Reallocated_Sector_Ct 0x0033 100 100 010 Pre-fail Always - 2 7 Seek_Error_Rate 0x000f 253 253 051 Pre-fail Always - 0 8 Seek_Time_Performance 0x0025 253 253 015 Pre-fail Offline - 0 9 Power_On_Hours 0x0032 100 100 000 Old_age Always - 2543 10 Spin_Retry_Count 0x0033 253 253 051 Pre-fail Always - 0 11 Calibration_Retry_Count 0x0012 253 253 000 Old_age Always - 0 12 Power_Cycle_Count 0x0032 100 100 000 Old_age Always - 721 13 Read_Soft_Error_Rate 0x000e 100 100 000 Old_age Always - 79115055 187 Unknown_Attribute 0x0032 253 253 000 Old_age Always - 0 188 Unknown_Attribute 0x0032 100 100 000 Old_age Always - 29 190 Unknown_Attribute 0x0022 071 049 000 Old_age Always - 29 194 Temperature_Celsius 0x0022 151 085 000 Old_age Always - 29 195 Hardware_ECC_Recovered 0x001a 100 100 000 Old_age Always - 79115055 196 Reallocated_Event_Count 0x0032 100 100 000 Old_age Always - 2 197 Current_Pending_Sector 0x0012 253 253 000 Old_age Always - 0 198 Offline_Uncorrectable 0x0030 253 253 000 Old_age Offline - 0 199 UDMA_CRC_Error_Count 0x003e 200 200 000 Old_age Always - 0 200 Multi_Zone_Error_Rate 0x000a 253 100 000 Old_age Always - 0 201 Soft_Read_Error_Rate 0x000a 100 100 000 Old_age Always - 0 202 TA_Increase_Count 0x0032 253 253 000 Old_age Always - 0 The email output (without mdcmd): his message is a status update for unRAID Tower ----------------------------------------------------------------- Server Name: Tower Status: Disk 2 Overheat! 151°C (DiskId: ata-SAMSUNG_HD501LJ) Date: Sun Sep 21 14:31:23 GMT 2008 Disk Temperature Status ----------------------------------------------------------------- Parity Disk [sdc]: Spun-Down (DiskId: ata-WDC_WD10EACS-00D6B0) Disk 1 [sdb]: Spun-Down (DiskId: ata-WDC_WD10EACS-00D6B0) Disk 2 [sdd]: 151°C (DiskId: ata-SAMSUNG_HD501LJ) Disk 3 [sde]: Spun-Down (DiskId: ata-WDC_WD4000YS-01MPB1) Disk 4 [sdf]: Spun-Down (DiskId: ata-WDC_WD4000YS-01MPB1) Disk SMART Health Status ----------------------------------------------------------------- Parity Disk Spun-Down (DiskId: ata-WDC_WD10EACS-00D6B0) Disk 1 Spun-Down (DiskId: ata-WDC_WD10EACS-00D6B0) Disk 2 PASSED (DiskId: ata-SAMSUNG_HD501LJ) Disk 3 Spun-Down (DiskId: ata-WDC_WD4000YS-01MPB1) Disk 4 Spun-Down (DiskId: ata-WDC_WD4000YS-01MPB1
September 21, 200817 yr I use the 10th field of the same line for my unmenu.awk script. You could use something like this: smartctl -d ata -A theDisk | grep -i temperature | awk '{ print $10 }' Joe L. I think I am able to modify the script according to the suggested method by Joe L, but maybe this 10th field should be used in general. Do I remember well, that wordpad can be used for editing, with proper unix ending?
September 21, 200817 yr You probably only need to change one line (line # 132 in the version I downloaded). from: # Disk is not spun down, try to get the disk temp diskTemp=`smartctl -d ata -A /dev/$1|awk '{if($2~"Temperature_Celsius"){print $4}}' 2>/dev/null` to: # Disk is not spun down, try to get the disk temp diskTemp=`smartctl -d ata -A /dev/$1|awk '{if($2~"Temperature_Celsius"){print $10}}' 2>/dev/null`
September 21, 200817 yr Author Olympia; Please try the attached. (edit: attachment removed - please refer to first post in this thread for the most recent version of unraid_notify)
September 21, 200817 yr Thanks Joe, I guessed that! And brainbone, thank you very much for this mod! This is something which seems to solve the spin down issues I always had, as a result it was controlled by only the disks' firmware, and I was never able to fully debug. Just one question: can be consider is at a safe solution? For example what happens with the data still in cache, not yet written to the disk, but there is a forced spin down. Is this data flushed somehow to the disk? Or I just misunderstand something, and this kind of situation cannot happen.
September 21, 200817 yr Author Did the temperature modification work? I believe the spin down should be as safe as allowing the firmware to control spin down, from the perspective of the OS caching write data.
September 21, 200817 yr Did the temperature modification work? Yes, works flawless now Thank you again! Was it OK in your setup?
September 21, 200817 yr Author Yes, worked OK on my system, but I have all Seagate drives. Refer to the first post in this thread - unraid_notify to reflect temperature fix. Thanks for your help!
September 21, 200817 yr Last two question: - Isn't the frequent (1 minute by default) smartctl check generate overload? Is it affect performance during heavy file transfer? - If I don't want to receive status messages, only alerts, than do I need to leave NotifyDelay empty?
September 21, 200817 yr Author - Isn't the frequent (1 minute by default) smartctl check generate overload? Is it affect performance during heavy file transfer? Didn't appear to effect performance on my system (Celeron 430 1.8ghz), even when I set it to check every 10 sec. If you notice any deterioration in performance, please let me know. - If I don't want to receive status messages, only alerts, than do I need to leave NotifyDelay empty? No. Leave "RcptTo" empty, or just comment it out (# RcptTo) Use only the "ErrorRcptTo" line instead.
September 30, 200817 yr I just got around to trying unraid_notify. Unfortunately I don't get emails. Here is the debug dialog. root@Tower:/boot/config# unraid_notify -d > < 220 sccmmhc92.asp.att.net - Maillennium ESMTP/MULTIBOX sccmmhc92 #40 > HELO Tower < 250 sccmmhc92.asp.att.net > MAIL FROM: <[email protected]> < 250 ok > RCPT TO: <[email protected]> < 250 ok; [simple] forward to <[email protected]> > DATA < 354 ok > . THEN AFTER SEVERAL MINUTES < 221 sccmmhc91.asp.att.net > QUIT Any ideas?
September 30, 200817 yr Author Could you post the contents of your unraid_notify.cfg (removing your smtp auth password for security, if you have one)?
September 30, 200817 yr Author Looking again at the debug output, it looks like your domain (vincit.com) is hosted at ProHostOne (guesing you're using "mail.vincit.com" as your smtp server), and your ISP (ATT or SBC?) is getting in the way and filtering your connection to mail.vincit.com on port 25. Unfortunately, it looks like ProHostOne does not provide an alternate port to 25 for SMTP connection, and I believe ATT/SBC only allow Secure (SSL) SMTP connections. I would contact your host to check if they have an alternate port for (Non SSL) SMTP connections (usually port 2525 or 587). If not, I would need to add Secure SMTP support, but this would require two additional packages: OpenSSL and Stunnel, increasing the ram and flash footprint of unraid_notify, at least for those needing SSL/Secure SMTP.
September 30, 200817 yr Author Who do you have for an ISP? I see "mail.mchsi.com" for your smtp server in unraid_notify.cfg, but in your debug output, sccmmhc92.asp.att.net is answering.
September 30, 200817 yr Author Never mind. mchsi.com = MediaCom = AT&T. See my post above. I believe your ISP requires Secure SMTP for sending mail.
October 1, 200817 yr Author Added SMTPS support using socat. Connecting to mail servers that require SSL support on port 465, like smtp.gmail.com, is now possible if you install the socat package. See top post for details.
October 1, 200817 yr can we remove the spindown stuff now that Tom has reverted 4.4 back to the "good" logic? Also, we have times below 1 minute now with 4.4. Ref the spindown time, is a blank entry a "disable" ?
October 1, 200817 yr The spin down changes are only in a beta, many users won't upgrade until it is out of beta. Perhaps it could be optional, or conditional on a version check?
October 2, 200817 yr Author Worked like a charm - thanks again! Good to hear! The spin down changes are only in a beta, many users won't upgrade until it is out of beta. Perhaps it could be optional, or conditional on a version check? I'll make it optional in the next version, but I'd like to wait to get more feedback on how 2.20 is working before pushing out a new one. can we remove the spindown stuff now that Tom has reverted 4.4 back to the "good" logic? Also, we have times below 1 minute now with 4.4. Ref the spindown time, is a blank entry a "disable" ? I'll put in an option to disable unraid_notify from doing a spindown. At the moment, a blank spindown time will default to 60min spindown. If unRAID spins down the drives before unraid_notify has a chance to, there shouldn't be any real problem or conflicts. unraid_notify will see that the drive is not spinning, and leave it alone. Below 1 min? I'm having trouble seeing where that is useful for anything but increased drive wear?
Archived
This topic is now archived and is closed to further replies.