Administrace komentářů

No a co by se s tím dalo udělat. OS se Centos 7, file system btrfs na LUKS oddílu. Patrně je to důsledek problému se SSD. Upgrade se firmware se na SSD nedá udělat, po koupi OCZ firmou Toshiba nejsou firmware dostupné. Provedl jsem dd SSD disku na HDD a systém spustil z něj. Provedl jsem odinstalaci a nainstalaci uvedeného balíku, nicméně odinstalace samozřejmě poškozené inody nezrušila. Takže jsem adresář munin přejmnoval a nová instalace balíku vše vytvořila správně. Munin porád nejede (což je patrně správně) protože jedna s akci kterou monitoruje je inode.

# su -s /bin/bash munin
bash-4.2$ /usr/share/munin/munin-update --debug --nofork 
2017/03/05 13:09:17 [DEBUG] Creating new lock file /var/run/munin/munin-update.lock
2017/03/05 13:09:17 [DEBUG] Creating lock : /var/run/munin/munin-update.lock succeeded
2017/03/05 13:09:17 [INFO]: Starting munin-update
2017/03/05 13:09:17 [DEBUG] Lock /var/run/munin/munin-localhost-localhost.lock already exists, checking process
2017/03/05 13:09:17 [DEBUG] Lock contained pid '30317'
2017/03/05 13:09:17 [INFO] Process 30317 is dead, stealing lock, removing file
2017/03/05 13:09:17 [DEBUG] Creating new lock file /var/run/munin/munin-localhost-localhost.lock
2017/03/05 13:09:17 [DEBUG] Creating lock : /var/run/munin/munin-localhost-localhost.lock succeeded
2017/03/05 13:09:17 [DEBUG] Reading state for localhost-localhost in /var/lib/munin/state-localhost-localhost.storable
2017/03/05 13:09:17 [INFO] starting work in 30343 for localhost/127.0.0.1:4949.
2017/03/05 13:09:17 [INFO] node localhost advertised itself as localhost.localdomain instead.
2017/03/05 13:09:17 TLS set to "disabled".
2017/03/05 13:09:17 [DEBUG] Negotiating capabilities
2017/03/05 13:09:17 [DEBUG] Writing to socket: "cap multigraph dirtyconfig
".
2017/03/05 13:09:17 [DEBUG] Node says /cap multigraph dirtyconfig/
2017/03/05 13:09:17 [DEBUG] Writing to socket: "list localhost.localdomain
".
2017/03/05 13:09:17 [DEBUG] for my open_inodes (open_inodes if_enp4s0 irqstats ntp_states df uptime swap load ntp_offset fw_conntrack cpu df_inode nfsd4 fw_forwarded_local open_files ntp_kernel_err forks diskstats memory vmstat if_enp6s1 fw_packets entropy ntp_kernel_pll_freq if_err_enp4s0 postfix_mailqueue processes users interrupts nfsd hddtemp_smartctl postfix_mailvolume proc_pri threads ntp_kernel_pll_off if_err_enp6s1)
2017/03/05 13:09:17 [DEBUG] Fetching service configuration for 'open_inodes'
2017/03/05 13:09:17 [DEBUG] Writing to socket: "config open_inodes
".
2017/03/05 13:09:17 [DEBUG] Reading from socket: "graph_title Inode table usage\ngraph_args --base 1000 -l 0\ngraph_vlabel number of open inodes\ngraph_category system\ngraph_info This graph monitors the Linux open inode table.\nused.label open inodes\nused.info The number of currently open inodes.\nmax.label inode table size\nmax.info The size of the system inode table. This is dynamically adjusted by the kernel.".
2017/03/05 13:09:17 [DEBUG] config: 0.075336 sec for 'open_inodes' on localhost/127.0.0.1/4949
2017/03/05 13:09:17 [DEBUG] Now parsing config output from plugin open_inodes on localhost
2017/03/05 13:09:17 [DEBUG] update_rate 0 for open_inodes on localhost/127.0.0.1:4949
2017/03/05 13:09:17 [DEBUG] No service data for open_inodes, fetching it
2017/03/05 13:09:17 [DEBUG] Writing to socket: "fetch open_inodes
".
2017/03/05 13:09:17 [DEBUG] data: 0.067993 sec for 'open_inodes' on localhost/127.0.0.1/4949
2017/03/05 13:09:17 [DEBUG] Now parsing fetch output from plugin open_inodes on localhost/127.0.0.1:4949
2017/03/05 13:09:17 [FETCH from open_inodes] used.value 73800
2017/03/05 13:09:17 [FETCH from open_inodes] Storing 73800 in used
2017/03/05 13:09:17 [FETCH from open_inodes] max.value 191015
2017/03/05 13:09:17 [FETCH from open_inodes] Storing 191015 in max
2017/03/05 13:09:17 [DEBUG] asking for a rrd of size : normal
2017/03/05 13:09:17 [DEBUG] rrd filename: /var/lib/munin/localhost/localhost-open_inodes-max-g.rrd
2017/03/05 13:09:17 [DEBUG] Updating /var/lib/munin/localhost/localhost-open_inodes-max-g.rrd with 1488715757:191015
Bus error
bash-4.2$ 

Porovnal jsem součty a chyby na FS jsou.

# btrfs scrub start -B /dev/dm-1
scrub done for d27abb02-b5d0-4eb3-86db-1b2d5a293fb2
        scrub started at Sun Mar  5 12:52:09 2017 and finished after 00:01:38
        total bytes scrubbed: 2.37GiB with 101 errors
        error details: verify=7 csum=94
        corrected errors: 0, uncorrectable errors: 101, unverified errors: 0
ERROR: there are uncorrectable errors

Takže teď jak to narovnat. Jak zjistit v jakých souborech mám chyby a případně je zrekonstruovat ze záloh.