How much do you need to know about disks to successfully manage a Linux system? What commands do what? How do you make good decisions about partitioning? What kind of troubleshooting tools are available? What kind of problems might you run into? This article covers a lot of territory – from looking into the basics of a Linux file systems to sampling some very useful commands.
Disk technology background
In the beginning days of Unix and later Linux, disks were physically large, but very small in terms of storage capacity. A 300 megabyte disk in the mid-90’s was the size of a shoebox. Today, you can get multi-terrabyte disks that are the size of a slice of toast.
Traditionally, files resided within file systems that resided in disk partitions that were themselves simply slices of disks. This organization still dominates today, though servers in large data centers often take on an entirely different structure.
/\ / \ / \ / file \ / \ /==========\ / \ / file system \ / \ /==================\ / disk partition \ /======================\ / disk \ /==========================\
This simplistic view still works for many systems, but these days there are lot of complexities that make disk management harder in some ways and easier in others. A file system might be virtual – no longer residing on a single disk and more complex to manage, but far easier to resize as needed. In fact, the entire system could be virtual. And what we might manage as if it were a single disk could actually be some portion of a very large disk array.
Disk management tasks
Sysadmins generally have to deal with many issues when it comes to managing disks. These include:
- Partitioning disks
- Creating file systems
- Mounting file systems
- Sharing file systems
- Monitoring free space within file systems
- Backing up (and sometimes restoring) file systems
The reasons to partition a disk include:
- protecting some file systems from running out of space (e.g., you may want the OS partition to be separated from home directories or applications to keep it from being affected if users’ files begin to take up far an excessive amount of disk space)
- improving performance
- allocating swap space
- facilitating maintenance and backups (e.g., you might be able to unmount /apps if it’s not part of / and you might want to back up /home more frequently than /usr)
- more efficient (and targeted) fsck
- maintaining (particularly on test systems) multiple operating systems
- reserving enough disk space for file system expansion
- sharing select file systems with other systems
For most Linux servers, partitioning is done before the servers are deployed. On the other hand, you might add disks at some later time or hold back some significant amount of free disk space for future use.
To make changes or verify partitions, enter a command such as fdisk /dev/sda to start fdisk interactively and then type m to see a list of the things that you can do with the fdisk command.
$ sudo fdisk /dev/sda Command (m for help): m Command action a toggle a bootable flag b edit bsd disklabel c toggle the dos compatibility flag d delete a partition l list known partition types m print this menu n add a new partition o create a new empty DOS partition table p print the partition table q quit without saving changes s create a new empty Sun disklabel t change a partition's system id u change display/entry units v verify the partition table w write table to disk and exit x extra functionality (experts only)
As you can see, the fdisk command provides a lot of functionality. The partitions that you set up may look something like this configuration in which four partitions have been set up on a single disk – /dev/sda.
sda +------------+------------------------+--------------------+------+ | / 40G | /home 80G | /apps 70G | swap | +------------+------------------------+--------------------+------+ sda1 sda2 sda3 sda4
Examining disk space and disk partitions
There are a number of excellent commands for examining disk partitions. The df command is one of the most commonly used commands for reporting on disk space usage. With the -h option, the df command displays the measurements in the most "human-friendly" format and that is, in fact, what the “h” is meant to imply. As you can see in the example below, the measurements are displayed in kilobytes, megabytes or gigabytes depending on the sizes rather than all using the same scale.
$ df -h Filesystem Size Used Avail Use% Mounted on udev 969M 4.0K 969M 1% /dev tmpfs 196M 1.1M 195M 1% /run /dev/sda1 37G 4.5G 31G 13% / none 4.0K 0 4.0K 0% /sys/fs/cgroup none 5.0M 0 5.0M 0% /run/lock none 980M 152K 979M 1% /run/shm none 100M 36K 100M 1% /run/user /dev/sda3 28G 44M 26G 1% /apps
The pydf command (think "python df" as it's really a python script) also provides a very useful disk usage display showing mount points and cute little illustrations for how full each partition is.
$ pydf Filesystem Size Used Avail Use% Mounted on /dev/sda1 37G 4534M 30G 12.1 [##...........] / /dev/sda3 27G 44M 26G 0.2 [.............] /apps
The parted command displays partition information in a different format:
$ sudo parted -l Model: ATA WDC WD800AAJS-60 (scsi) Disk /dev/sda: 80.0GB Sector size (logical/physical): 512B/512B Partition Table: msdos Number Start End Size Type File system Flags 1 1049kB 40.0GB 40.0GB primary ext4 boot 2 40.0GB 50.0GB 10.0GB primary linux-swap(v1) 3 50.0GB 80.0GB 30.0GB primary ext4
The lsblk (list block devices) command illustrates the relationship between disks and their partitions graphically and also supplies the major and minor device numbers and mount points.
$ lsblk NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT sda 8:0 0 74.5G 0 disk ├─sda1 8:1 0 37.3G 0 part / ├─sda2 8:2 0 9.3G 0 part [SWAP] └─sda3 8:3 0 28G 0 part /apps
The fdisk command reports more details on disk partitions and uses very different numbers. You can also use fdisk to create or delete partitions, list unpartitioned space, change a partition type, or verify the partition table.
$ sudo fdisk -l Disk /dev/sda: 80.0 GB, 80026361856 bytes 255 heads, 63 sectors/track, 9729 cylinders, total 156301488 sectors Units = sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disk identifier: 0x000f114b Device Boot Start End Blocks Id System /dev/sda1 * 2048 78125055 39061504 83 Linux /dev/sda2 78125056 97656831 9765888 82 Linux swap / Solaris /dev/sda3 97656832 156301311 29322240 83 Linux
The sfdisk command is similar to fdisk, but makes some partition manipulation activities easier to perform.
$ sudo sfdisk -l -uM Disk /dev/sda: 9729 cylinders, 255 heads, 63 sectors/track Units = mebibytes of 1048576 bytes, blocks of 1024 bytes, counting from 0 Device Boot Start End MiB #blocks Id System /dev/sda1 * 1 38146 38146 39061504 83 Linux /dev/sda2 38147 47683 9537 9765888 82 Linux swap / Solaris /dev/sda3 47684 76318 28635 29322240 83 Linux /dev/sda4 0 - 0 0 0 Empty
NOTE: A mebibyte (MiB) = 220 bytes or 1,048,576 bytes.
The cfdisk command can also be used to display or manipulate disk partitions.
$ sudo cfdisk cfdisk (util-linux 2.20.1) Disk Drive: /dev/sda Size: 80026361856 bytes, 80.0 GB Heads: 255 Sectors per Track: 63 Cylinders: 9729 Name Flags Part Type FS Type [Label] Size (MB) -------------------------------------------------------------------------- Pri/Log Free Space 1.05* sda1 Boot Primary ext4 39998.99* sda2 Primary swap 10000.27* sda3 Primary ext4 30025.98* Pri/Log Free Space 0.10* [ Help ] [ New ] [ Print ] [ Quit ] [ Units ] [ Write ] Create new partition from free space
Monitoring disk performance
The iostat command can display statistics that illustrate how disks are performing, including how heavily they are being used. It also displays important measurements that show how busy the CPU is and how much of its resources are used for types of work. The system described below is idle more then 95% of the time. More importantly for our focus on disks, the %iowait (CPU waiting on disk IO) is very low. This would not be true if the disk were unusually busy and disk IO were a bottleneck.
$ iostat -x 60 Linux 3.13.0-129-generic (stinkbug) 08/31/2017 _x86_64_ (2 CPU) avg-cpu: %user %nice %system %iowait %steal %idle 0.93 1.15 0.35 1.86 0.00 95.73 Device: rrqm/s wrqm/s r/s w/s rkB/s wkB/s avgrq-sz avgqu-sz await r_await w_await svctm %util sda 8.37 3.26 13.41 2.79 341.14 191.82 65.79 0.61 37.60 30.40 72.14 2.52 4.08
Probably one of the most informative commands for looking at disk health is smartctl (part of smartmontools). While the command generates a lot of output, it provides valuable measurements that might help you pinpoint disk problems, particularly once you get used to working with its extensive output.
$ sudo smartctl -a /dev/sda1 smartctl 6.2 2013-07-26 r3841 [x86_64-linux-3.13.0-129-generic] (local build) Copyright (C) 2002-13, Bruce Allen, Christian Franke, www.smartmontools.org === START OF INFORMATION SECTION === Model Family: Western Digital Caviar Blue Serial ATA Device Model: WDC WD800AAJS-60M0A0 Serial Number: WD-WMAV37134378 LU WWN Device Id: 5 0014ee 0015c85ef Firmware Version: 02.03E02 User Capacity: 80,026,361,856 bytes [80.0 GB] Sector Size: 512 bytes logical/physical Device is: In smartctl database [for details use: -P show] ATA Version is: ATA8-ACS (minor revision not indicated) SATA Version is: SATA 2.5, 3.0 Gb/s Local Time is: Thu Aug 31 15:30:19 2017 EDT SMART support is: Available - device has SMART capability. SMART support is: Enabled === START OF READ SMART DATA SECTION === SMART overall-health self-assessment test result: PASSED General SMART Values: Offline data collection status: (0x82) Offline data collection activity was completed without error. Auto Offline Data Collection: Enabled. Self-test execution status: ( 0) The previous self-test routine completed without error or no self-test has ever been run. Total time to complete Offline data collection: ( 2700) seconds. Offline data collection capabilities: (0x5b) SMART execute Offline immediate. Auto Offline data collection on/off support. Suspend Offline collection upon new command. Offline surface scan supported. Self-test supported. No Conveyance Self-test supported. Selective Self-test supported. SMART capabilities: (0x0003) Saves SMART data before entering power-saving mode. Supports SMART auto save timer. Error logging capability: (0x01) Error logging supported. General Purpose Logging supported. Short self-test routine recommended polling time: ( 2) minutes. Extended self-test routine recommended polling time: ( 36) minutes. SCT capabilities: (0x303f) SCT Status supported. SCT Error Recovery Control supported. SCT Feature Control supported. SCT Data Table supported. 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 0x002f 200 200 051 Pre-fail Always - 0 3 Spin_Up_Time 0x0027 143 140 021 Pre-fail Always - 3841 4 Start_Stop_Count 0x0032 100 100 000 Old_age Always - 178 5 Reallocated_Sector_Ct 0x0033 200 200 140 Pre-fail Always - 0 7 Seek_Error_Rate 0x002f 100 253 051 Pre-fail Always - 0 9 Power_On_Hours 0x0032 058 058 000 Old_age Always - 31203 10 Spin_Retry_Count 0x0033 100 100 051 Pre-fail Always - 0 11 Calibration_Retry_Count 0x0032 100 100 000 Old_age Always - 0 12 Power_Cycle_Count 0x0032 100 100 000 Old_age Always - 175 184 End-to-End_Error 0x0033 100 100 097 Pre-fail Always - 0 187 Reported_Uncorrect 0x0032 100 100 000 Old_age Always - 0 188 Command_Timeout 0x0032 100 100 000 Old_age Always - 0 190 Airflow_Temperature_Cel 0x0022 066 062 040 Old_age Always - 34 192 Power-Off_Retract_Count 0x0032 200 200 000 Old_age Always - 103 193 Load_Cycle_Count 0x0032 200 200 000 Old_age Always - 178 196 Reallocated_Event_Count 0x0032 200 200 000 Old_age Always - 0 197 Current_Pending_Sector 0x0032 200 200 000 Old_age Always - 0 198 Offline_Uncorrectable 0x0030 200 200 000 Old_age Offline - 0 199 UDMA_CRC_Error_Count 0x0032 200 200 000 Old_age Always - 0 200 Multi_Zone_Error_Rate 0x0008 200 200 000 Old_age Offline - 0 SMART Error Log Version: 1 No Errors Logged SMART Self-test log structure revision number 1 Num Test_Description Status Remaining LifeTime(hours) LBA_of_first_error # 1 Short offline Completed without error 00% 30349 - # 2 Extended offline Aborted by host 80% 0 - SMART Selective self-test log data structure revision number 1 SPAN MIN_LBA MAX_LBA CURRENT_TEST_STATUS 1 0 0 Not_testing 2 0 0 Not_testing 3 0 0 Not_testing 4 0 0 Not_testing 5 0 0 Not_testing Selective self-test flags (0x0): After scanning selected spans, do NOT read-scan remainder of disk. If Selective self-test is pending on power-up, resume after 0 minute delay.
There are numerous other commands for examining disks and file systems. Those described here are some of the most useful and informative. Using them periodically has advantages as the easiest way to spot problems is becoming so used to the output of commands such as these that you easily spot the kind of differences that might indicate problems.