How to control the filesystem's write-back cache avoiding some such issue?
by kcynice from LinuxQuestions.org on (#5D83E)
Here, I have a large tar file with gzip compression, about 50GB(Im not sure its original size, 100GB or a little larger). I want to extract it into a ext4 partition of a mobile SSD disk, however, I will always fail with write back error and accompanied by some IO error:
PHP Code:mce: CPU8: Package temperature/speed normal
[ 1197.106933] usb 2-1: USB disconnect, device number 2
[ 1197.107180] usb 2-1: cmd cmplt err -108
[ 1197.107181] usb 2-1: cmd cmplt err -108
[ 1197.107181] usb 2-1: cmd cmplt err -108
[ 1197.107182] usb 2-1: cmd cmplt err -108
[ 1197.107182] usb 2-1: cmd cmplt err -108
[ 1197.107183] usb 2-1: cmd cmplt err -108
[ 1197.107183] usb 2-1: cmd cmplt err -108
[ 1197.107184] xhci_hcd 0000:00:14.0: WARN Set TR Deq Ptr cmd failed due to incorrect slot or ep state.
[ 1197.107190] sd 1:0:0:0: [sdb] tag#17 uas_zap_pending 0 uas-tag 1 inflight: CMD
[ 1197.107191] sd 1:0:0:0: [sdb] tag#17 CDB: opcode=0x2a 2a 00 14 dc 7a ff 00 00 10 00
[ 1197.107192] sd 1:0:0:0: [sdb] tag#18 uas_zap_pending 0 uas-tag 2 inflight: CMD
[ 1197.107193] sd 1:0:0:0: [sdb] tag#18 CDB: opcode=0x2a 2a 00 14 dc 7b 0f 00 04 00 00
[ 1197.107194] sd 1:0:0:0: [sdb] tag#19 uas_zap_pending 0 uas-tag 3 inflight: CMD
[ 1197.107194] sd 1:0:0:0: [sdb] tag#19 CDB: opcode=0x2a 2a 00 14 dc 7f 0f 00 04 00 00
[ 1197.107195] sd 1:0:0:0: [sdb] tag#16 uas_zap_pending 0 uas-tag 4 inflight: CMD
[ 1197.107195] sd 1:0:0:0: [sdb] tag#16 CDB: opcode=0x2a 2a 00 14 dc 83 0f 00 01 70 00
[ 1197.107196] sd 1:0:0:0: [sdb] tag#23 uas_zap_pending 0 uas-tag 5 inflight: CMD
[ 1197.107197] sd 1:0:0:0: [sdb] tag#23 CDB: opcode=0x2a 2a 00 15 19 3f 8f 00 00 a0 00
[ 1197.107197] sd 1:0:0:0: [sdb] tag#20 uas_zap_pending 0 uas-tag 6 inflight: CMD
[ 1197.107198] sd 1:0:0:0: [sdb] tag#20 CDB: opcode=0x2a 2a 00 15 58 ff c7 00 00 28 00
[ 1197.107199] sd 1:0:0:0: [sdb] tag#21 uas_zap_pending 0 uas-tag 7 inflight: CMD
[ 1197.107199] sd 1:0:0:0: [sdb] tag#21 CDB: opcode=0x2a 2a 00 15 19 3f 5f 00 00 10 00
[ 1197.107208] sd 1:0:0:0: [sdb] tag#17 UNKNOWN(0x2003) Result: hostbyte=0x01 driverbyte=0x00
[ 1197.107209] sd 1:0:0:0: [sdb] tag#17 CDB: opcode=0x2a 2a 00 14 dc 7a ff 00 00 10 00
[ 1197.107210] blk_update_request: I/O error, dev sdb, sector 349993727 op 0x1:(WRITE) flags 0x0 phys_seg 2 prio class 0
[ 1197.107212] EXT4-fs warning (device sdb2): ext4_end_bio:309: I/O error 10 writing to inode 10898572 (offset 0 size 8192 starting block 43749217)
[ 1197.107213] Buffer I/O error on device sdb2, logical block 43609954
[ 1197.107215] Buffer I/O error on device sdb2, logical block 43609955
[ 1197.107219] sd 1:0:0:0: [sdb] tag#18 UNKNOWN(0x2003) Result: hostbyte=0x01 driverbyte=0x00
[ 1197.107220] sd 1:0:0:0: [sdb] tag#18 CDB: opcode=0x2a 2a 00 14 dc 7b 0f 00 04 00 00
[ 1197.107220] blk_update_request: I/O error, dev sdb, sector 349993743 op 0x1:(WRITE) flags 0x4000 phys_seg 128 prio class 0
[ 1197.107222] sd 1:0:0:0: [sdb] tag#19 UNKNOWN(0x2003) Result: hostbyte=0x01 driverbyte=0x00
[ 1197.107223] sd 1:0:0:0: [sdb] tag#19 CDB: opcode=0x2a 2a 00 14 dc 7f 0f 00 04 00 00
[ 1197.107223] blk_update_request: I/O error, dev sdb, sector 349994767 op 0x1:(WRITE) flags 0x4000 phys_seg 128 prio class 0
[ 1197.107225] sd 1:0:0:0: [sdb] tag#16 UNKNOWN(0x2003) Result: hostbyte=0x01 driverbyte=0x00
[ 1197.107225] sd 1:0:0:0: [sdb] tag#16 CDB: opcode=0x2a 2a 00 14 dc 83 0f 00 01 70 00
[ 1197.107226] blk_update_request: I/O error, dev sdb, sector 349995791 op 0x1:(WRITE) flags 0x0 phys_seg 46 prio class 0
[ 1197.107227] EXT4-fs warning (device sdb2): ext4_end_bio:309: I/O error 10 writing to inode 10898573 (offset 0 size 1212416 starting block 43749513)
[ 1197.107227] Buffer I/O error on device sdb2, logical block 43609956
[ 1197.107228] Buffer I/O error on device sdb2, logical block 43609957
[ 1197.107228] Buffer I/O error on device sdb2, logical block 43609958
[ 1197.107229] Buffer I/O error on device sdb2, logical block 43609959
[ 1197.107230] Buffer I/O error on device sdb2, logical block 43609960
[ 1197.107230] Buffer I/O error on device sdb2, logical block 43609961
[ 1197.107231] Buffer I/O error on device sdb2, logical block 43609962
[ 1197.107231] Buffer I/O error on device sdb2, logical block 43609963
[ 1197.107279] EXT4-fs warning (device sdb2): ext4_end_bio:309: I/O error 10 writing to inode 10898574 (offset 0 size 24576 starting block 43749519)
[ 1197.107283] sd 1:0:0:0: [sdb] tag#23 UNKNOWN(0x2003) Result: hostbyte=0x01 driverbyte=0x00
[ 1197.107284] sd 1:0:0:0: [sdb] tag#23 CDB: opcode=0x2a 2a 00 15 19 3f 8f 00 00 a0 00
[ 1197.107284] blk_update_request: I/O error, dev sdb, sector 353976207 op 0x1:(WRITE) flags 0x0 phys_seg 20 prio class 0
[ 1197.107285] EXT4-fs warning (device sdb2): ext4_end_bio:309: I/O error 10 writing to inode 11027277 (offset 0 size 12288 starting block 44247028)
[ 1197.107286] EXT4-fs warning (device sdb2): ext4_end_bio:309: I/O error 10 writing to inode 11027278 (offset 0 size 4096 starting block 44247029)
[ 1197.107287] EXT4-fs warning (device sdb2): ext4_end_bio:309: I/O error 10 writing to inode 11027283 (offset 0 size 4096 starting block 44247030)
[ 1197.107288] EXT4-fs warning (device sdb2): ext4_end_bio:309: I/O error 10 writing to inode 11027285 (offset 0 size 4096 starting block 44247031)
[ 1197.107289] EXT4-fs warning (device sdb2): ext4_end_bio:309: I/O error 10 writing to inode 11027286 (offset 0 size 4096 starting block 44247032)
[ 1197.107291] EXT4-fs warning (device sdb2): ext4_end_bio:309: I/O error 10 writing to inode 11027288 (offset 0 size 4096 starting block 44247033)
[ 1197.107292] EXT4-fs warning (device sdb2): ext4_end_bio:309: I/O error 10 writing to inode 11027290 (offset 0 size 12288 starting block 44247036)
[ 1197.107298] sd 1:0:0:0: [sdb] tag#20 UNKNOWN(0x2003) Result: hostbyte=0x01 driverbyte=0x00
[ 1197.107298] sd 1:0:0:0: [sdb] tag#20 CDB: opcode=0x2a 2a 00 15 58 ff c7 00 00 28 00
[ 1197.107299] blk_update_request: I/O error, dev sdb, sector 358154183 op 0x1:(WRITE) flags 0x0 phys_seg 5 prio class 0
[ 1197.107303] sd 1:0:0:0: [sdb] tag#21 UNKNOWN(0x2003) Result: hostbyte=0x01 driverbyte=0x00
[ 1197.107303] sd 1:0:0:0: [sdb] tag#21 CDB: opcode=0x2a 2a 00 15 19 3f 5f 00 00 10 00
[ 1197.107304] blk_update_request: I/O error, dev sdb, sector 353976159 op 0x1:(WRITE) flags 0x0 phys_seg 2 prio class 0
[ 1197.107307] blk_update_request: I/O error, dev sdb, sector 358161351 op 0x1:(WRITE) flags 0x0 phys_seg 20 prio class 0
[ 1197.107319] blk_update_request: I/O error, dev sdb, sector 362120703 op 0x1:(WRITE) flags 0x0 phys_seg 6 prio class 0
[ 1197.107323] blk_update_request: I/O error, dev sdb, sector 362121495 op 0x1:(WRITE) flags 0x0 phys_seg 5 prio class 0
[ 1197.154204] EXT4-fs error (device sdb2): ext4_wait_block_bitmap:517: comm kworker/u24:3: Cannot read block bitmap - block_group = 1570, block_bitmap = 51380226
[ 1197.154214] Buffer I/O error on dev sdb2, logical block 0, lost sync page write
[ 1197.154217] EXT4-fs (sdb2): I/O error while writing superblock
[ 1197.157278] JBD2: Detected IO errors while flushing file data on sdb2-8
[ 1197.157355] Aborting journal on device sdb2-8.
[ 1197.157358] EXT4-fs error (device sdb2) in ext4_reserve_inode_write:5941: Journal has aborted
[ 1197.157360] Buffer I/O error on dev sdb2, logical block 38830080, lost sync page write
[ 1197.157362] JBD2: Error -5 detected when updating journal superblock for sdb2-8.
[ 1197.157363] Buffer I/O error on dev sdb2, logical block 0, lost sync page write
[ 1197.157365] EXT4-fs (sdb2): I/O error while writing superblock
[ 1197.157379] EXT4-fs error (device sdb2): mpage_map_and_submit_extent:2589: comm kworker/u24:3: Failed to mark inode 12848840 dirty
[ 1197.157381] Buffer I/O error on dev sdb2, logical block 0, lost sync page write
[ 1197.157381] EXT4-fs (sdb2): I/O error while writing superblock
[ 1197.157382] EXT4-fs error (device sdb2) in ext4_writepages:2901: Journal has aborted
[ 1197.157384] Buffer I/O error on dev sdb2, logical block 0, lost sync page write
[ 1197.157385] EXT4-fs (sdb2): I/O error while writing superblock
[ 1197.157406] Buffer I/O error on dev sdb2, logical block 0, lost sync page write
[ 1197.157407] EXT4-fs (sdb2): I/O error while writing superblock
[ 1197.157407] EXT4-fs error (device sdb2): ext4_journal_check_start:61: Detected aborted journal
[ 1197.157408] EXT4-fs (sdb2): Remounting filesystem read-only
[ 1197.157409] Buffer I/O error on dev sdb2, logical block 0, lost sync page write
[ 1197.157410] EXT4-fs (sdb2): I/O error while writing superblock
[ 1197.157411] EXT4-fs (sdb2): ext4_writepages: jbd2_start: 53248 pages, ino 12848842; err -30
[ 1197.157672] JBD2: Detected IO errors while flushing file data on sdb2-8
so, I re-format the partition with jfs, and got the same error. Then, I have to re-mount the parition with sync flag, and wait it more than 24 hours without error, but, I can wait more time, because it just only completed about one of the third decompression job.
To getting enough reason to believe the physical disk is OK, I tested the physical disk partition with dd command and got none error. I have no idea to avoiding such error so far. Any hints will be appreciated.
Thanks, all
PHP Code:mce: CPU8: Package temperature/speed normal
[ 1197.106933] usb 2-1: USB disconnect, device number 2
[ 1197.107180] usb 2-1: cmd cmplt err -108
[ 1197.107181] usb 2-1: cmd cmplt err -108
[ 1197.107181] usb 2-1: cmd cmplt err -108
[ 1197.107182] usb 2-1: cmd cmplt err -108
[ 1197.107182] usb 2-1: cmd cmplt err -108
[ 1197.107183] usb 2-1: cmd cmplt err -108
[ 1197.107183] usb 2-1: cmd cmplt err -108
[ 1197.107184] xhci_hcd 0000:00:14.0: WARN Set TR Deq Ptr cmd failed due to incorrect slot or ep state.
[ 1197.107190] sd 1:0:0:0: [sdb] tag#17 uas_zap_pending 0 uas-tag 1 inflight: CMD
[ 1197.107191] sd 1:0:0:0: [sdb] tag#17 CDB: opcode=0x2a 2a 00 14 dc 7a ff 00 00 10 00
[ 1197.107192] sd 1:0:0:0: [sdb] tag#18 uas_zap_pending 0 uas-tag 2 inflight: CMD
[ 1197.107193] sd 1:0:0:0: [sdb] tag#18 CDB: opcode=0x2a 2a 00 14 dc 7b 0f 00 04 00 00
[ 1197.107194] sd 1:0:0:0: [sdb] tag#19 uas_zap_pending 0 uas-tag 3 inflight: CMD
[ 1197.107194] sd 1:0:0:0: [sdb] tag#19 CDB: opcode=0x2a 2a 00 14 dc 7f 0f 00 04 00 00
[ 1197.107195] sd 1:0:0:0: [sdb] tag#16 uas_zap_pending 0 uas-tag 4 inflight: CMD
[ 1197.107195] sd 1:0:0:0: [sdb] tag#16 CDB: opcode=0x2a 2a 00 14 dc 83 0f 00 01 70 00
[ 1197.107196] sd 1:0:0:0: [sdb] tag#23 uas_zap_pending 0 uas-tag 5 inflight: CMD
[ 1197.107197] sd 1:0:0:0: [sdb] tag#23 CDB: opcode=0x2a 2a 00 15 19 3f 8f 00 00 a0 00
[ 1197.107197] sd 1:0:0:0: [sdb] tag#20 uas_zap_pending 0 uas-tag 6 inflight: CMD
[ 1197.107198] sd 1:0:0:0: [sdb] tag#20 CDB: opcode=0x2a 2a 00 15 58 ff c7 00 00 28 00
[ 1197.107199] sd 1:0:0:0: [sdb] tag#21 uas_zap_pending 0 uas-tag 7 inflight: CMD
[ 1197.107199] sd 1:0:0:0: [sdb] tag#21 CDB: opcode=0x2a 2a 00 15 19 3f 5f 00 00 10 00
[ 1197.107208] sd 1:0:0:0: [sdb] tag#17 UNKNOWN(0x2003) Result: hostbyte=0x01 driverbyte=0x00
[ 1197.107209] sd 1:0:0:0: [sdb] tag#17 CDB: opcode=0x2a 2a 00 14 dc 7a ff 00 00 10 00
[ 1197.107210] blk_update_request: I/O error, dev sdb, sector 349993727 op 0x1:(WRITE) flags 0x0 phys_seg 2 prio class 0
[ 1197.107212] EXT4-fs warning (device sdb2): ext4_end_bio:309: I/O error 10 writing to inode 10898572 (offset 0 size 8192 starting block 43749217)
[ 1197.107213] Buffer I/O error on device sdb2, logical block 43609954
[ 1197.107215] Buffer I/O error on device sdb2, logical block 43609955
[ 1197.107219] sd 1:0:0:0: [sdb] tag#18 UNKNOWN(0x2003) Result: hostbyte=0x01 driverbyte=0x00
[ 1197.107220] sd 1:0:0:0: [sdb] tag#18 CDB: opcode=0x2a 2a 00 14 dc 7b 0f 00 04 00 00
[ 1197.107220] blk_update_request: I/O error, dev sdb, sector 349993743 op 0x1:(WRITE) flags 0x4000 phys_seg 128 prio class 0
[ 1197.107222] sd 1:0:0:0: [sdb] tag#19 UNKNOWN(0x2003) Result: hostbyte=0x01 driverbyte=0x00
[ 1197.107223] sd 1:0:0:0: [sdb] tag#19 CDB: opcode=0x2a 2a 00 14 dc 7f 0f 00 04 00 00
[ 1197.107223] blk_update_request: I/O error, dev sdb, sector 349994767 op 0x1:(WRITE) flags 0x4000 phys_seg 128 prio class 0
[ 1197.107225] sd 1:0:0:0: [sdb] tag#16 UNKNOWN(0x2003) Result: hostbyte=0x01 driverbyte=0x00
[ 1197.107225] sd 1:0:0:0: [sdb] tag#16 CDB: opcode=0x2a 2a 00 14 dc 83 0f 00 01 70 00
[ 1197.107226] blk_update_request: I/O error, dev sdb, sector 349995791 op 0x1:(WRITE) flags 0x0 phys_seg 46 prio class 0
[ 1197.107227] EXT4-fs warning (device sdb2): ext4_end_bio:309: I/O error 10 writing to inode 10898573 (offset 0 size 1212416 starting block 43749513)
[ 1197.107227] Buffer I/O error on device sdb2, logical block 43609956
[ 1197.107228] Buffer I/O error on device sdb2, logical block 43609957
[ 1197.107228] Buffer I/O error on device sdb2, logical block 43609958
[ 1197.107229] Buffer I/O error on device sdb2, logical block 43609959
[ 1197.107230] Buffer I/O error on device sdb2, logical block 43609960
[ 1197.107230] Buffer I/O error on device sdb2, logical block 43609961
[ 1197.107231] Buffer I/O error on device sdb2, logical block 43609962
[ 1197.107231] Buffer I/O error on device sdb2, logical block 43609963
[ 1197.107279] EXT4-fs warning (device sdb2): ext4_end_bio:309: I/O error 10 writing to inode 10898574 (offset 0 size 24576 starting block 43749519)
[ 1197.107283] sd 1:0:0:0: [sdb] tag#23 UNKNOWN(0x2003) Result: hostbyte=0x01 driverbyte=0x00
[ 1197.107284] sd 1:0:0:0: [sdb] tag#23 CDB: opcode=0x2a 2a 00 15 19 3f 8f 00 00 a0 00
[ 1197.107284] blk_update_request: I/O error, dev sdb, sector 353976207 op 0x1:(WRITE) flags 0x0 phys_seg 20 prio class 0
[ 1197.107285] EXT4-fs warning (device sdb2): ext4_end_bio:309: I/O error 10 writing to inode 11027277 (offset 0 size 12288 starting block 44247028)
[ 1197.107286] EXT4-fs warning (device sdb2): ext4_end_bio:309: I/O error 10 writing to inode 11027278 (offset 0 size 4096 starting block 44247029)
[ 1197.107287] EXT4-fs warning (device sdb2): ext4_end_bio:309: I/O error 10 writing to inode 11027283 (offset 0 size 4096 starting block 44247030)
[ 1197.107288] EXT4-fs warning (device sdb2): ext4_end_bio:309: I/O error 10 writing to inode 11027285 (offset 0 size 4096 starting block 44247031)
[ 1197.107289] EXT4-fs warning (device sdb2): ext4_end_bio:309: I/O error 10 writing to inode 11027286 (offset 0 size 4096 starting block 44247032)
[ 1197.107291] EXT4-fs warning (device sdb2): ext4_end_bio:309: I/O error 10 writing to inode 11027288 (offset 0 size 4096 starting block 44247033)
[ 1197.107292] EXT4-fs warning (device sdb2): ext4_end_bio:309: I/O error 10 writing to inode 11027290 (offset 0 size 12288 starting block 44247036)
[ 1197.107298] sd 1:0:0:0: [sdb] tag#20 UNKNOWN(0x2003) Result: hostbyte=0x01 driverbyte=0x00
[ 1197.107298] sd 1:0:0:0: [sdb] tag#20 CDB: opcode=0x2a 2a 00 15 58 ff c7 00 00 28 00
[ 1197.107299] blk_update_request: I/O error, dev sdb, sector 358154183 op 0x1:(WRITE) flags 0x0 phys_seg 5 prio class 0
[ 1197.107303] sd 1:0:0:0: [sdb] tag#21 UNKNOWN(0x2003) Result: hostbyte=0x01 driverbyte=0x00
[ 1197.107303] sd 1:0:0:0: [sdb] tag#21 CDB: opcode=0x2a 2a 00 15 19 3f 5f 00 00 10 00
[ 1197.107304] blk_update_request: I/O error, dev sdb, sector 353976159 op 0x1:(WRITE) flags 0x0 phys_seg 2 prio class 0
[ 1197.107307] blk_update_request: I/O error, dev sdb, sector 358161351 op 0x1:(WRITE) flags 0x0 phys_seg 20 prio class 0
[ 1197.107319] blk_update_request: I/O error, dev sdb, sector 362120703 op 0x1:(WRITE) flags 0x0 phys_seg 6 prio class 0
[ 1197.107323] blk_update_request: I/O error, dev sdb, sector 362121495 op 0x1:(WRITE) flags 0x0 phys_seg 5 prio class 0
[ 1197.154204] EXT4-fs error (device sdb2): ext4_wait_block_bitmap:517: comm kworker/u24:3: Cannot read block bitmap - block_group = 1570, block_bitmap = 51380226
[ 1197.154214] Buffer I/O error on dev sdb2, logical block 0, lost sync page write
[ 1197.154217] EXT4-fs (sdb2): I/O error while writing superblock
[ 1197.157278] JBD2: Detected IO errors while flushing file data on sdb2-8
[ 1197.157355] Aborting journal on device sdb2-8.
[ 1197.157358] EXT4-fs error (device sdb2) in ext4_reserve_inode_write:5941: Journal has aborted
[ 1197.157360] Buffer I/O error on dev sdb2, logical block 38830080, lost sync page write
[ 1197.157362] JBD2: Error -5 detected when updating journal superblock for sdb2-8.
[ 1197.157363] Buffer I/O error on dev sdb2, logical block 0, lost sync page write
[ 1197.157365] EXT4-fs (sdb2): I/O error while writing superblock
[ 1197.157379] EXT4-fs error (device sdb2): mpage_map_and_submit_extent:2589: comm kworker/u24:3: Failed to mark inode 12848840 dirty
[ 1197.157381] Buffer I/O error on dev sdb2, logical block 0, lost sync page write
[ 1197.157381] EXT4-fs (sdb2): I/O error while writing superblock
[ 1197.157382] EXT4-fs error (device sdb2) in ext4_writepages:2901: Journal has aborted
[ 1197.157384] Buffer I/O error on dev sdb2, logical block 0, lost sync page write
[ 1197.157385] EXT4-fs (sdb2): I/O error while writing superblock
[ 1197.157406] Buffer I/O error on dev sdb2, logical block 0, lost sync page write
[ 1197.157407] EXT4-fs (sdb2): I/O error while writing superblock
[ 1197.157407] EXT4-fs error (device sdb2): ext4_journal_check_start:61: Detected aborted journal
[ 1197.157408] EXT4-fs (sdb2): Remounting filesystem read-only
[ 1197.157409] Buffer I/O error on dev sdb2, logical block 0, lost sync page write
[ 1197.157410] EXT4-fs (sdb2): I/O error while writing superblock
[ 1197.157411] EXT4-fs (sdb2): ext4_writepages: jbd2_start: 53248 pages, ino 12848842; err -30
[ 1197.157672] JBD2: Detected IO errors while flushing file data on sdb2-8
so, I re-format the partition with jfs, and got the same error. Then, I have to re-mount the parition with sync flag, and wait it more than 24 hours without error, but, I can wait more time, because it just only completed about one of the third decompression job.
To getting enough reason to believe the physical disk is OK, I tested the physical disk partition with dd command and got none error. I have no idea to avoiding such error so far. Any hints will be appreciated.
Thanks, all