1 \chapter{Testing Your Tape Drive With Bacula}
2 \label{TapeTestingChapter}
3 \index[general]{Testing Your Tape Drive With Bacula}
5 This chapter is concerned with testing and configuring your tape drive to make
6 sure that it will work properly with Bacula using the {\bf btape} program.
9 \section{Get Your Tape Drive Working}
11 In general, you should follow the following steps to get your tape drive to
12 work with Bacula. Start with a tape mounted in your drive. If you have an
13 autochanger, load a tape into the drive. We use {\bf /dev/nst0} as the tape
14 drive name, you will need to adapt it according to your system.
16 Do not proceed to the next item until you have succeeded with the previous
20 \item Make sure that Bacula (the Storage daemon) is not running
21 or that you have {\bf unmount}ed the drive you will use
24 \item Use tar to write to, then read from your drive:
28 mt -f /dev/nst0 rewind
30 mt -f /dev/nst0 rewind
36 \item Make sure you have a valid and correct Device resource corresponding
37 to your drive. For Linux users, generally, the default one works. For
38 FreeBSD users, there are two possible Device configurations (see below).
39 For other drives and/or OSes, you will need to first ensure that your
40 system tape modes are properly setup (see below), then possibly modify
41 you Device resource depending on the output from the btape program (next
42 item). When doing this, you should consult the \bsysxrlink{Storage Daemon
43 Configuration}{StoredConfChapter}{main}{chapter} of the \mainman{}.
45 \item If you are using a Fibre Channel to connect your tape drive to
46 Bacula, please be sure to disable any caching in the NSR (network
47 storage router, which is a Fibre Channel to SCSI converter).
49 \item Run the btape {\bf test} command:
53 ./btape -c bacula-sd.conf /dev/nst0
59 It isn't necessary to run the autochanger part of the test at this time,
60 but do not go past this point until the basic test succeeds. If you do
61 have an autochanger, please be sure to read the \bsysxrlink{Autochanger
62 chapter}{AutochangersChapter}{main}{chapter} of the \mainman{}.
64 \item Run the btape {\bf fill} command, preferably with two volumes. This
65 can take a long time. If you have an autochanger and it is configured, Bacula
66 will automatically use it. If you do not have it configured, you can manually
67 issue the appropriate {\bf mtx} command, or press the autochanger buttons to
68 change the tape when requested to do so.
70 \item FreeBSD users, if you have a pre-5.0 system run the {\bf tapetest}
71 program, and make sure your system is patched if necessary. The tapetest
72 program can be found in the platform/freebsd directory. The instructions
73 for its use are at the top of the file.
75 \item Run Bacula, and backup a reasonably small directory, say 60
76 Megabytes. Do three successive backups of this directory.
78 \item Stop Bacula, then restart it. Do another full backup of the same
79 directory. Then stop and restart Bacula.
81 \item Do a restore of the directory backed up, by entering the following
82 restore command, being careful to restore it to an alternate location:
87 restore select all done
93 Do a {\bf diff} on the restored directory to ensure it is identical to the
94 original directory. If you are going to backup multiple different systems
95 (Linux, Windows, Mac, Solaris, FreeBSD, ...), be sure you test the restore
98 \item If you have an autochanger, you should now go back to the btape program
99 and run the autochanger test:
103 ./btape -c bacula-sd.conf /dev/nst0
109 Adjust your autochanger as necessary to ensure that it works correctly. See
110 the Autochanger chapter of this manual for a complete discussion of testing
113 \item We strongly recommend that you use a dedicated SCSI
114 controller for your tape drives. Scanners are known to induce
115 serious problems with the SCSI bus, causing it to reset. If the
116 SCSI bus is reset while Bacula has the tape drive open, it will
117 most likely be fatal to your tape since the drive will rewind.
118 These kinds of problems show up in the system log. For example,
119 the following was most likely caused by a scanner:
123 Feb 14 17:29:55 epohost kernel: (scsi0:A:2:0): No or incomplete CDB sent to device.
124 Feb 14 17:29:55 epohost kernel: scsi0: Issued Channel A Bus Reset. 1 SCBs aborted
130 If you have reached this point, you stand a good chance of having everything
131 work. If you get into trouble at any point, {\bf carefully} read the
132 documentation given below. If you cannot get past some point, ask the {\bf
133 bacula-users} email list, but specify which of the steps you have successfully
134 completed. In particular, you may want to look at the
135 \ilink{Tips for Resolving Problems}{problems1} section below.
138 \label{NoTapeInDrive}
139 \subsection{Problems When no Tape in Drive}
140 \index[general]{Problems When no Tape in Drive}
141 When Bacula was first written the Linux 2.4 kernel permitted opening the
142 drive whether or not there was a tape in the drive. Thus the Bacula code is
143 based on the concept that if the drive cannot be opened, there is a serious
144 problem, and the job is failed.
146 With version 2.6 of the Linux kernel, if there is no tape in the drive, the
147 OS will wait two minutes (default) and then return a failure, and consequently,
148 Bacula version 1.36 and below will fail the job. This is important to keep
149 in mind, because if you use an option such as {\bf Offline on Unmount =
150 yes}, there will be a point when there is no tape in the drive, and if
151 another job starts or if Bacula asks the operator to mount a tape, when
152 Bacula attempts to open the drive (about a 20 minute delay), it will fail
153 and Bacula will fail the job.
155 In version 1.38.x, the Bacula code partially gets around this problem -- at
156 least in the initial open of the drive. However, functions like Polling
157 the drive do not work correctly if there is no tape in the drive.
158 Providing you do not use {\bf Offline on Unmount = yes}, you should not
159 experience job failures as mentioned above. If you do experience such
160 failures, you can also increase the {\bf Maximum Open Wait} time interval,
161 which will give you more time to mount the next tape before the job is
164 \subsection{Specifying the Configuration File}
165 \index[general]{File!Specifying the Configuration}
166 \index[general]{Specifying the Configuration File}
168 Starting with version 1.27, each of the tape utility programs including the
169 {\bf btape} program requires a valid Storage daemon configuration file
170 (actually, the only part of the configuration file that {\bf btape} needs is
171 the {\bf Device} resource definitions). This permits {\bf btape} to find the
172 configuration parameters for your archive device (generally a tape drive).
173 Without those parameters, the testing and utility programs do not know how to
174 properly read and write your drive. By default, they use {\bf bacula-sd.conf}
175 in the current directory, but you may specify a different configuration file
176 using the {\bf -c} option.
178 \subsection{Specifying a Device Name For a Tape}
179 \index[general]{Tape!Specifying a Device Name For a}
180 \index[general]{Specifying a Device Name For a Tape}
182 {\bf btape} {\bf device-name} where the Volume can be found. In the case of a
183 tape, this is the physical device name such as {\bf /dev/nst0} or {\bf
184 /dev/rmt/0ubn} depending on your system that you specify on the Archive Device
185 directive. For the program to work, it must find the identical name in the
186 Device resource of the configuration file. If the name is not found in the
187 list of physical names, the utility program will compare the name you entered
188 to the Device names (rather than the Archive device names).
190 When specifying a tape device, it is preferable that the "non-rewind"
191 variant of the device file name be given. In addition, on systems such as
192 Sun, which have multiple tape access methods, you must be sure to specify
193 to use Berkeley I/O conventions with the device. The
194 {\bf b} in the Solaris (Sun) archive specification {\bf /dev/rmt/0mbn} is
195 what is needed in this case. Bacula does not support SysV tape drive
198 See below for specifying Volume names.
200 \subsection{Specifying a Device Name For a File}
201 \index[general]{File!Specifying a Device Name For a}
202 \index[general]{Specifying a Device Name For a File}
204 If you are attempting to read or write an archive file rather than a tape, the
205 {\bf device-name} should be the full path to the archive location including
206 the filename. The filename (last part of the specification) will be stripped
207 and used as the Volume name, and the path (first part before the filename)
208 must have the same entry in the configuration file. So, the path is equivalent
209 to the archive device name, and the filename is equivalent to the volume name.
214 \index[general]{Btape}
216 This program permits a number of elementary tape operations via a tty command
217 interface. The {\bf test} command, described below, can be very useful for
218 testing tape drive compatibility problems. Aside from initial testing of tape
219 drive compatibility with {\bf Bacula}, {\bf btape} will be mostly used by
220 developers writing new tape drivers.
222 {\bf btape} can be dangerous to use with existing {\bf Bacula} tapes because
223 it will relabel a tape or write on the tape if so requested regardless of
224 whether or not the tape contains valuable data, so please be careful and use
225 it only on blank tapes.
227 To work properly, {\bf btape} needs to read the Storage daemon's configuration
228 file. As a default, it will look for {\bf bacula-sd.conf} in the current
229 directory. If your configuration file is elsewhere, please use the {\bf -c}
230 option to specify where.
232 The physical device name or the Device resource name must be specified on the
233 command line, and this same device name must be present in the Storage
234 daemon's configuration file read by {\bf btape}
238 Usage: btape [options] device_name
239 -b <file> specify bootstrap file
240 -c <file> set configuration file to file
241 -d <nn> set debug level to nn
242 -p proceed inspite of I/O errors
245 -? print this message.
249 \subsection{Using btape to Verify your Tape Drive}
250 \index[general]{Using btape to Verify your Tape Drive}
251 \index[general]{Drive!Using btape to Verify your Tape}
253 An important reason for this program is to ensure that a Storage daemon
254 configuration file is defined so that Bacula will correctly read and write
257 It is highly recommended that you run the {\bf test} command before running
258 your first Bacula job to ensure that the parameters you have defined for your
259 storage device (tape drive) will permit {\bf Bacula} to function properly. You
260 only need to mount a blank tape, enter the command, and the output should be
261 reasonably self explanatory. For example:
265 (ensure that Bacula is not running)
266 ./btape -c /usr/bin/bacula/bacula-sd.conf /dev/nst0
274 Tape block granularity is 1024 bytes.
275 btape: btape.c:376 Using device: /dev/nst0
280 Enter the test command:
288 The output produced should be something similar to the following: I've cut the
289 listing short because it is frequently updated to have new tests.
293 === Append files test ===
294 This test is essential to Bacula.
295 I'm going to write one record in file 0,
296 two records in file 1,
297 and three records in file 2
298 btape: btape.c:387 Rewound /dev/nst0
299 btape: btape.c:855 Wrote one record of 64412 bytes.
300 btape: btape.c:857 Wrote block to device.
301 btape: btape.c:410 Wrote EOF to /dev/nst0
302 btape: btape.c:855 Wrote one record of 64412 bytes.
303 btape: btape.c:857 Wrote block to device.
304 btape: btape.c:855 Wrote one record of 64412 bytes.
305 btape: btape.c:857 Wrote block to device.
306 btape: btape.c:410 Wrote EOF to /dev/nst0
307 btape: btape.c:855 Wrote one record of 64412 bytes.
308 btape: btape.c:857 Wrote block to device.
309 btape: btape.c:855 Wrote one record of 64412 bytes.
310 btape: btape.c:857 Wrote block to device.
311 btape: btape.c:855 Wrote one record of 64412 bytes.
312 btape: btape.c:857 Wrote block to device.
313 btape: btape.c:410 Wrote EOF to /dev/nst0
314 btape: btape.c:387 Rewound /dev/nst0
315 btape: btape.c:693 Now moving to end of media.
316 btape: btape.c:427 Moved to end of media
317 We should be in file 3. I am at file 3. This is correct!
318 Now the important part, I am going to attempt to append to the tape.
320 === End Append files test ===
324 If you do not successfully complete the above test, please resolve the
325 problem(s) before attempting to use {\bf Bacula}. Depending on your tape
326 drive, the test may recommend that you add certain records to your
327 configuration. We strongly recommend that you do so and then re-run the above
328 test to insure it works the first time.
330 Some of the suggestions it provides for resolving the problems may or may not
331 be useful. If at all possible avoid using fixed blocking. If the test suddenly
332 starts to print a long series of:
342 then almost certainly, you are running your drive in fixed block mode rather
343 than variable block mode. See below for more help of resolving fix
344 versus variable block problems.
346 It is also possible that you have your drive
347 set in SysV tape drive mode. The drive must use BSD tape conventions.
348 See the section above on setting your {\bf Archive device} correctly.
350 For FreeBSD users, please see the notes below for doing further testing of
353 \subsection{Testing tape drive speed}
354 \label{sec:btapespeed}
356 To determine the best configuration of your tape drive, you can run the
357 \texttt{speed} command available in the \texttt{btape} program.
359 This command can have the following arguments:
361 \item[\texttt{file\_size=n}] Specify the Maximum File Size for this test
362 (between 1 and 5GB). This counter is in GB.
363 \item[\texttt{nb\_file=n}] Specify the number of file to be written. The amount
364 of data should be greater than your memory ($file\_size*nb\_file$).
365 \item[\texttt{skip\_zero}] This flag permits to skip tests with constant
367 \item[\texttt{skip\_random}] This flag permits to skip tests with random
369 \item[\texttt{skip\_raw}] This flag permits to skip tests with raw access.
370 \item[\texttt{skip\_block}] This flag permits to skip tests with Bacula block
375 *speed file_size=3 skip_raw
376 btape.c:1078 Test with zero data and bacula block structure.
377 btape.c:956 Begin writing 3 files of 3.221 GB with blocks of 129024 bytes.
378 ++++++++++++++++++++++++++++++++++++++++++
379 btape.c:604 Wrote 1 EOF to "Drive-0" (/dev/nst0)
380 btape.c:406 Volume bytes=3.221 GB. Write rate = 44.128 MB/s
382 btape.c:383 Total Volume bytes=9.664 GB. Total Write rate = 43.531 MB/s
384 btape.c:1090 Test with random data, should give the minimum throughput.
385 btape.c:956 Begin writing 3 files of 3.221 GB with blocks of 129024 bytes.
386 +++++++++++++++++++++++++++++++++++++++++++
387 btape.c:604 Wrote 1 EOF to "Drive-0" (/dev/nst0)
388 btape.c:406 Volume bytes=3.221 GB. Write rate = 7.271 MB/s
389 +++++++++++++++++++++++++++++++++++++++++++
391 btape.c:383 Total Volume bytes=9.664 GB. Total Write rate = 7.365 MB/s
395 When using compression, the random test will give your the minimum throughput
396 of your drive . The test using constant string will give you the maximum speed
397 of your hardware chain. (cpu, memory, scsi card, cable, drive, tape).
399 You can change the block size in the Storage Daemon configuration file.
402 \subsection{Linux SCSI Tricks}
403 \index[general]{Tricks!Linux SCSI}
404 \index[general]{Linux SCSI Tricks}
406 You can find out what SCSI devices you have by doing:
418 [0:0:0:0] disk ATA ST3160812AS 3.AD /dev/sda
419 [2:0:4:0] tape HP Ultrium 2-SCSI F6CH /dev/st0
420 [2:0:5:0] tape HP Ultrium 2-SCSI F6CH /dev/st1
421 [2:0:6:0] mediumx OVERLAND LXB 0107 -
422 [2:0:9:0] tape HP Ultrium 1-SCSI E50H /dev/st2
423 [2:0:10:0] mediumx OVERLAND LXB 0107 -
427 There are two drives in one autochanger: /dev/st0 and /dev/st1
428 and a third tape drive at /dev/st2. For using them with Bacula, one
429 would normally reference them as /dev/nst0 ... /dev/nst2. Not also,
430 there are two different autochangers identified as "mediumx OVERLAND LXB".
431 They can be addressed via their /dev/sgN designation, which can be
432 obtained by counting from the beginning as 0 to each changer. In the
433 above case, the two changers are located on /dev/sg3 and /dev/sg5. The one
434 at /dev/sg3, controls drives /dev/nst0 and /dev/nst1; and the one at
435 /dev/sg5 controles drive /dev/nst2.
437 If you do not have the {\bf lsscsi} command, you can obtain the same
438 information as follows:
446 For the above example with the three drives and two autochangers,
452 Host: scsi0 Channel: 00 Id: 00 Lun: 00
453 Vendor: ATA Model: ST3160812AS Rev: 3.AD
454 Type: Direct-Access ANSI SCSI revision: 05
455 Host: scsi2 Channel: 00 Id: 04 Lun: 00
456 Vendor: HP Model: Ultrium 2-SCSI Rev: F6CH
457 Type: Sequential-Access ANSI SCSI revision: 03
458 Host: scsi2 Channel: 00 Id: 05 Lun: 00
459 Vendor: HP Model: Ultrium 2-SCSI Rev: F6CH
460 Type: Sequential-Access ANSI SCSI revision: 03
461 Host: scsi2 Channel: 00 Id: 06 Lun: 00
462 Vendor: OVERLAND Model: LXB Rev: 0107
463 Type: Medium Changer ANSI SCSI revision: 02
464 Host: scsi2 Channel: 00 Id: 09 Lun: 00
465 Vendor: HP Model: Ultrium 1-SCSI Rev: E50H
466 Type: Sequential-Access ANSI SCSI revision: 03
467 Host: scsi2 Channel: 00 Id: 10 Lun: 00
468 Vendor: OVERLAND Model: LXB Rev: 0107
469 Type: Medium Changer ANSI SCSI revision: 02
474 As an additional example, I get the following (on a different machine from the
480 Host: scsi2 Channel: 00 Id: 01 Lun: 00
481 Vendor: HP Model: C5713A Rev: H107
482 Type: Sequential-Access ANSI SCSI revision: 02
483 Host: scsi2 Channel: 00 Id: 04 Lun: 00
484 Vendor: SONY Model: SDT-10000 Rev: 0110
485 Type: Sequential-Access ANSI SCSI revision: 02
489 The above represents first an autochanger and second a simple
490 tape drive. The HP changer (the first entry) uses the same SCSI channel
491 for data and for control, so in Bacula, you would use:
494 Archive Device = /dev/nst0
495 Changer Device = /dev/sg0
499 If you want to remove the SDT-10000 device, you can do so as root with:
503 echo "scsi remove-single-device 2 0 4 0">/proc/scsi/scsi
507 and you can put add it back with:
511 echo "scsi add-single-device 2 0 4 0">/proc/scsi/scsi
515 where the 2 0 4 0 are the Host, Channel, Id, and Lun as seen on the output
516 from {\bf cat /proc/scsi/scsi}. Note, the Channel must be specified as
519 Below is a slightly more complicated output, which is a single autochanger
520 with two drives, and which operates the changer on a different channel
521 from from the drives:
526 Host: scsi0 Channel: 00 Id: 00 Lun: 00
527 Vendor: ATA Model: WDC WD1600JD-75H Rev: 08.0
528 Type: Direct-Access ANSI SCSI revision: 05
529 Host: scsi2 Channel: 00 Id: 04 Lun: 00
530 Vendor: HP Model: Ultrium 2-SCSI Rev: F6CH
531 Type: Sequential-Access ANSI SCSI revision: 03
532 Host: scsi2 Channel: 00 Id: 05 Lun: 00
533 Vendor: HP Model: Ultrium 2-SCSI Rev: F6CH
534 Type: Sequential-Access ANSI SCSI revision: 03
535 Host: scsi2 Channel: 00 Id: 06 Lun: 00
536 Vendor: OVERLAND Model: LXB Rev: 0106
537 Type: Medium Changer ANSI SCSI revision: 02
541 The above tape drives are accessed on /dev/nst0 and /dev/nst1, while
542 the control channel for those two drives is /dev/sg3.
547 \section{Tips for Resolving Problems}
548 \index[general]{Problems!Tips for Resolving}
549 \index[general]{Tips for Resolving Problems}
551 \label{CannotRestore}
552 \subsection{Bacula Saves But Cannot Restore Files}
553 \index[general]{Files!Bacula Saves But Cannot Restore}
554 \index[general]{Bacula Saves But Cannot Restore Files}
556 If you are getting error messages such as:
560 Volume data error at 0:1! Wanted block-id: "BB02", got "". Buffer discarded
564 It is very likely that Bacula has tried to do block positioning and ended up
565 at an invalid block. This can happen if your tape drive is in fixed block mode
566 while Bacula's default is variable blocks. Note that in such cases, Bacula is
567 perfectly able to write to your Volumes (tapes), but cannot position to read
570 There are two possible solutions.
573 \item The first and best is to always ensure that your drive is in variable
574 block mode. Note, it can switch back to fixed block mode on a reboot or if
575 another program uses the drive. So on such systems you need to modify the
576 Bacula startup files to explicitly set:
580 mt -f /dev/nst0 defblksize 0
584 or whatever is appropriate on your system. Note, if you are running a Linux
585 system, and the above command does not work, it is most likely because you
586 have not loaded the appropriate {\bf mt} package, which is often called
587 {\bf mt\_st}, but may differ according to your distribution.
589 \item The second possibility, especially, if Bacula wrote while the drive was
590 in fixed block mode, is to turn off block positioning in Bacula. This is done
595 Block Positioning = no
599 to the Device resource. This is not the recommended procedure because it can
600 enormously slow down recovery of files, but it may help where all else
601 fails. This directive is available in version 1.35.5 or later (and not yet
605 If you are getting error messages such as:
608 Volume data error at 0:0!
609 Block checksum mismatch in block=0 len=32625 calc=345678 blk=123456
613 You are getting tape read errors, and this is most likely due to
614 one of the following things:
616 \item An old or bad tape.
617 \item A dirty drive that needs cleaning (particularly for DDS drives).
618 \item A loose SCSI cable.
619 \item Old firmware in your drive. Make sure you have the latest firmware
621 \item Computer memory errors.
622 \item Over-clocking your CPU.
623 \item A bad SCSI card.
628 \subsection{Bacula Cannot Open the Device}
629 \index[general]{Device!Bacula Cannot Open the}
630 \index[general]{Bacula Cannot Open the Device}
632 If you get an error message such as:
636 dev open failed: dev.c:265 stored: unable to open
637 device /dev/nst0:> ERR=No such device or address
641 the first time you run a job, it is most likely due to the fact that you
642 specified the incorrect device name on your {\bf Archive Device}.
644 If Bacula works fine with your drive, then all off a sudden you get error
645 messages similar to the one shown above, it is quite possible that your driver
646 module is being removed because the kernel deems it idle. This is done via
647 {\bf crontab} with the use of {\bf rmmod -a}. To fix the problem, you can
648 remove this entry from {\bf crontab}, or you can manually {\bf modprob} your
649 driver module (or add it to the local startup script). Thanks to Alan Brown
651 \label{IncorrectFiles}
653 \subsection{Incorrect File Number}
654 \index[general]{Number!Incorrect File}
655 \index[general]{Incorrect File Number}
657 When Bacula moves to the end of the medium, it normally uses the {\bf
658 ioctl(MTEOM)} function. Then Bacula uses the {\bf ioctl(MTIOCGET)} function to
659 retrieve the current file position from the {\bf mt\_fileno} field. Some SCSI
660 tape drivers will use a fast means of seeking to the end of the medium and in
661 doing so, they will not know the current file position and hence return a {\bf
662 -1}. As a consequence, if you get {\bf "This is NOT correct!"} in the
663 positioning tests, this may be the cause. You must correct this condition in
664 order for Bacula to work.
666 There are two possible solutions to the above problem of incorrect file
670 \item Figure out how to configure your SCSI driver to keep track of the file
671 position during the MTEOM request. This is the preferred solution.
672 \item Modify the {\bf Device} resource of your {\bf bacula-sd.conf} file to
677 Hardware End of File = no
681 This will cause Bacula to use the MTFSF request to seek to the end of the
682 medium, and Bacula will keep track of the file number itself.
685 \label{IncorrectBlocks}
686 \subsection{Incorrect Number of Blocks or Positioning Errors}
687 \index[general]{Testing!Incorrect Number of Blocks or Positioning Errors}
688 \index[general]{Incorrect Number of Blocks or Positioning Errors}
690 {\bf Bacula's} preferred method of working with tape drives (sequential
691 devices) is to run in variable block mode, and this is what is set by default.
692 You should first ensure that your tape drive is set for variable block mode
695 If your tape drive is in fixed block mode and you have told Bacula to use
696 different fixed block sizes or variable block sizes (default), you will get
697 errors when Bacula attempts to forward space to the correct block (the kernel
698 driver's idea of tape blocks will not correspond to Bacula's).
700 All modern tape drives support variable tape blocks, but some older drives (in
701 particular the QIC drives) as well as the ATAPI ide-scsi driver run only in
702 fixed block mode. The Travan tape drives also apparently must run in fixed
703 block mode (to be confirmed).
705 Even in variable block mode, with the exception of the first record on the
706 second or subsequent volume of a multi-volume backup, Bacula will write blocks
707 of a fixed size. However, in reading a tape, Bacula will assume that for each
708 read request, exactly one block from the tape will be transferred. This the
709 most common way that tape drives work and is well supported by {\bf Bacula}.
711 Drives that run in fixed block mode can cause serious problems for Bacula if
712 the drive's block size does not correspond exactly to {\bf Bacula's} block
713 size. In fixed block size mode, drivers may transmit a partial block or
714 multiple blocks for a single read request. From {\bf Bacula's} point of view,
715 this destroys the concept of tape blocks. It is much better to run in variable
716 block mode, and almost all modern drives (the OnStream is an exception) run in
717 variable block mode. In order for Bacula to run in fixed block mode, you must
718 include the following records in the Storage daemon's Device resource
723 Minimum Block Size = nnn
724 Maximum Block Size = nnn
728 where {\bf nnn} must be the same for both records and must be identical to the
729 driver's fixed block size.
731 We recommend that you avoid this configuration if at all possible by using
732 variable block sizes.
734 If you must run with fixed size blocks, make sure they are not 512 bytes. This
735 is too small and the overhead that Bacula has with each record will become
736 excessive. If at all possible set any fixed block size to something like
737 64,512 bytes or possibly 32,768 if 64,512 is too large for your drive. See
738 below for the details on checking and setting the default drive block size.
740 To recover files from tapes written in fixed block mode, see below.
743 \subsection{Ensuring that the Tape Modes Are Properly Set -- {\bf Linux
745 \index[general]{Ensuring that the Tape Modes Are Properly Set -- Linux Only}
747 If you have a modern SCSI tape drive and you are having problems with the {\bf
748 test} command as noted above, it may be that some program has set one or more
749 of your SCSI driver's options to non-default values. For example, if your
750 driver is set to work in SysV manner, Bacula will not work correctly because
751 it expects BSD behavior. To reset your tape drive to the default values, you
752 can try the following, but {\bf ONLY} if you have a SCSI tape drive on a {\bf
758 mt -f /dev/nst0 rewind
759 mt -f /dev/nst0 stoptions buffer-writes async-writes read-ahead
763 The above commands will clear all options and then set those specified. None
764 of the specified options are required by Bacula, but a number of other options
765 such as SysV behavior must not be set. Bacula does not support SysV tape
766 behavior. On systems other than Linux, you will need to consult your {\bf mt}
767 man pages or documentation to figure out how to do the same thing. This should
768 not really be necessary though -- for example, on both Linux and Solaris
769 systems, the default tape driver options are compatible with Bacula.
770 On Solaris systems, you must take care to specify the correct device
771 name on the {\bf Archive device} directive. See above for more details.
773 You may also want to ensure that no prior program has set the default block
774 size, as happened to one user, by explicitly turning it off with:
778 mt -f /dev/nst0 defblksize 0
782 If you are running a Linux
783 system, and the above command does not work, it is most likely because you
784 have not loaded the appropriate {\bf mt} package, which is often called
785 {\bf mt\_st}, but may differ according to your distribution.
787 If you would like to know what options you have set before making any of the
788 changes noted above, you can now view them on Linux systems, thanks to a tip
789 provided by Willem Riede. Do the following:
794 mt -f /dev/nst0 stsetoptions 0
795 grep st0 /var/log/messages
799 and you will get output that looks something like the following:
803 kernel: st0: Mode 0 options: buffer writes: 1, async writes: 1, read ahead: 1
804 kernel: st0: can bsr: 0, two FMs: 0, fast mteom: 0, auto lock: 0,
805 kernel: st0: defs for wr: 0, no block limits: 0, partitions: 0, s2 log: 0
806 kernel: st0: sysv: 0 nowait: 0
810 Note, I have chopped off the beginning of the line with the date and machine
811 name for presentation purposes.
813 Some people find that the above settings only last until the next reboot, so
814 please check this otherwise you may have unexpected problems.
816 Beginning with Bacula version 1.35.8, if Bacula detects that you are running
817 in variable block mode, it will attempt to set your drive appropriately. All
818 OSes permit setting variable block mode, but some OSes do not permit setting
819 the other modes that Bacula needs to function properly.
822 \subsection{Tape Hardware Compression and Blocking Size}
823 \index[general]{Tape Hardware Compression and Blocking Size}
824 \index[general]{Size!Tape Hardware Compression and Blocking Size}
826 You should be able to verify the tape compression status with sysfs on Linux.
828 cat /sys/class/scsi_tape/nst0/default_compression
831 You can, turn it on by using (on Linux):
836 mt -f /dev/nst0 defcompression 1
840 and of course, if you use a zero instead of the one at the end, you will turn
843 If you have built the {\bf mtx} program in the {\bf depkgs} package, you can
844 use tapeinfo to get quite a bit of information about your tape drive even if
845 it is not an autochanger. This program is called using the SCSI control
846 device. On Linux for tape drive /dev/nst0, this is usually /dev/sg0, while on
847 FreeBSD for /dev/nsa0, the control device is often /dev/pass2. For example on
848 my DDS-4 drive (/dev/nst0), I get the following:
853 Product Type: Tape Drive
855 Product ID: 'C5713A '
864 Medium Type: Not Loaded
870 where the {\bf DataCompEnabled: yes} means that tape hardware compression is
871 turned on. You can turn it on and off (yes|no) by using the {\bf mt}
872 commands given above. Also, this output will tell you if the {\bf BlockSize}
873 is non-zero and hence set for a particular block size. Bacula is not likely to
874 work in such a situation because it will normally attempt to write blocks of
875 64,512 bytes, except the last block of the job which will generally be
876 shorter. The first thing to try is setting the default block size to zero
877 using the {\bf mt -f /dev/nst0 defblksize 0} command as shown above.
878 On FreeBSD, this would be something like: {\bf mt -f /dev/nsa0 blocksize 0}.
880 On some operating systems with some tape drives, the amount of data that
881 can be written to the tape and whether or not compression is enabled is
882 determined by the density usually the {\bf mt -f /dev/nst0 setdensity xxx} command.
883 Often {\bf mt -f /dev/nst0 status} will print out the current
884 density code that is used with the drive. Most systems, but unfortunately
885 not all, set the density to the maximum by default. On some systems, you
886 can also get a list of all available density codes with:
887 {\bf mt -f /dev/nst0 densities} or a similar {\bf mt} command.
888 Note, for DLT and SDLT devices, no-compression versus compression is very
889 often controlled by the density code. On FreeBSD systems, the compression
890 mode is set using {\bf mt -f /dev/nsa0 comp xxx} where xxx is the
891 mode you want. In general, see {\bf man mt} for the options available on
894 Note, some of the above {\bf mt} commands may not be persistent depending
895 on your system configuration. That is they may be reset if a program
896 other than Bacula uses the drive or, as is frequently the case, on reboot
899 If your tape drive requires fixed block sizes (very unusual), you can use the
904 Minimum Block Size = nnn
905 Maximum Block Size = nnn
909 in your Storage daemon's Device resource to force Bacula to write fixed size
910 blocks (where you sent nnn to be the same for both of the above records). This
911 should be done only if your drive does not support variable block sizes, or
912 you have some other strong reasons for using fixed block sizes. As mentioned
913 above, a small fixed block size of 512 or 1024 bytes will be very inefficient.
914 Try to set any fixed block size to something like 64,512 bytes or larger if
915 your drive will support it.
917 Also, note that the {\bf Medium Type} field of the output of {\bf tapeinfo}
918 reports {\bf Not Loaded}, which is not correct. As a consequence, you should
919 ignore that field as well as the {\bf Attached Changer} field.
921 To recover files from tapes written in fixed block mode, see below.
924 \subsection{Tape Modes on FreeBSD}
925 \index[general]{FreeBSD!Tape Modes on}
926 \index[general]{Tape Modes on FreeBSD}
928 On most FreeBSD systems such as 4.9 and most tape drives, Bacula should run
933 mt -f /dev/nsa0 seteotmodel 2
934 mt -f /dev/nsa0 blocksize 0
935 mt -f /dev/nsa0 comp enable
939 You might want to put those commands in a startup script to make sure your
940 tape driver is properly initialized before running Bacula, because
941 depending on your system configuration, these modes may be reset if a
942 program other than Bacula uses the drive or when your system is rebooted.
944 Then according to what the {\bf btape test} command returns, you will probably
945 need to set the following (see below for an alternative):
949 Hardware End of Medium = no
951 Backward Space Record = no
952 Backward Space File = no
953 Fast Forward Space File = no
958 Then be sure to run some append tests with Bacula where you start and stop
959 Bacula between appending to the tape, or use {\bf btape} version 1.35.1 or
960 greater, which includes simulation of stopping/restarting Bacula.
962 Please see the file {\bf platforms/freebsd/pthreads-fix.txt} in the main
963 Bacula directory concerning {\bf important} information concerning
964 compatibility of Bacula and your system. A much more optimal Device
965 configuration is shown below, but does not work with all tape drives. Please
966 test carefully before putting either into production.
968 Note, for FreeBSD 4.10-RELEASE, using a Sony TSL11000 L100 DDS4 with an
969 autochanger set to variable block size and DCLZ compression, Brian McDonald
970 reports that to get Bacula to append correctly between Bacula executions,
971 the correct values to use are:
975 mt -f /dev/nsa0 seteotmodel 1
976 mt -f /dev/nsa0 blocksize 0
977 mt -f /dev/nsa0 comp enable
985 Hardware End of Medium = no
987 Backward Space Record = no
988 Backward Space File = no
989 Fast Forward Space File = yes
994 This has been confirmed by several other people using different hardware. This
995 configuration is the preferred one because it uses one EOF and no backspacing
996 at the end of the tape, which works much more efficiently and reliably with
999 Finally, here is a Device configuration that Danny Butroyd reports to work
1000 correctly with the Overland Powerloader tape library using LT0-2 and
1005 # Overland Powerloader LT02 - 17 slots single drive
1009 Archive Device = /dev/nsa0
1010 AutomaticMount = yes;
1012 RemovableMedia = yes;
1014 Changer Command = "/usr/local/sbin/mtx-changer %c %o %S %a %d"
1015 Changer Device = /dev/pass2
1017 Alert Command = "sh -c 'tapeinfo -f %c |grep TapeAlert|cat'"
1019 # FreeBSD Specific Settings
1020 Offline On Unmount = no
1021 Hardware End of Medium = no
1023 Backward Space Record = no
1024 Fast Forward Space File = no
1028 The following Device resource works fine with Dell PowerVault 110T and
1029 120T devices on both FreeBSD 5.3 and on NetBSD 3.0. It also works
1030 with Sony AIT-2 drives on FreeBSD.
1035 # FreeBSD/NetBSD Specific Settings
1036 Hardware End of Medium = no
1038 Backward Space Record = no
1039 Fast Forward Space File = yes
1045 On FreeBSD version 6.0, it is reported that you can even set
1046 Backward Space Record = yes.
1050 \subsection{Finding your Tape Drives and Autochangers on FreeBSD}
1051 \index[general]{FreeBSD!Finding Tape Drives and Autochangers}
1052 \index[general]{Finding Tape Drives and Autochangers on FreeBSD}
1054 On FreeBSD, you can do a {\bf camcontrol devlist} as root to determine what
1055 drives and autochangers you have. For example,
1059 undef# camcontrol devlist
1060 at scbus0 target 2 lun 0 (pass0,sa0)
1061 at scbus0 target 4 lun 0 (pass1,sa1)
1062 at scbus0 target 4 lun 1 (pass2)
1066 from the above, you can determine that there is a tape drive on {\bf /dev/sa0}
1067 and another on {\bf /dev/sa1} in addition since there is a second line for the
1068 drive on {\bf /dev/sa1}, you know can assume that it is the control device for
1069 the autochanger (i.e. {\bf /dev/pass2}). It is also the control device name to
1070 use when invoking the tapeinfo program. E.g.
1074 tapeinfo -f /dev/pass2
1080 \subsection{Using the OnStream driver on Linux Systems}
1081 \index[general]{Using the OnStream driver on Linux Systems}
1082 \index[general]{Systems!Using the OnStream driver on Linux}
1084 Bacula version 1.33 (not 1.32x) is now working and ready for testing with the
1085 OnStream kernel osst driver version 0.9.14 or above. Osst is available from:
1086 \elink{http://sourceforge.net/projects/osst/}
1087 {http://sourceforge.net/projects/osst/}.
1089 To make Bacula work you must first load the new driver then, as root, do:
1093 mt -f /dev/nosst0 defblksize 32768
1097 Also you must add the following to your Device resource in your Storage
1102 Minimum Block Size = 32768
1103 Maximum Block Size = 32768
1107 Here is a Device specification provided by Michel Meyers that is known to
1113 Name = "Onstream DI-30"
1114 Media Type = "ADR-30"
1115 Archive Device = /dev/nosst0
1116 Minimum Block Size = 32768
1117 Maximum Block Size = 32768
1118 Hardware End of Medium = yes
1120 Backward Space File = yes
1121 Fast Forward Space File = yes
1123 AutomaticMount = yes
1125 Removable Media = yes
1130 \section{Hardware Compression on EXB-8900}
1131 \index[general]{Hardware Compression on EXB-8900}
1132 \index[general]{EXB-8900!Hardware Compression}
1134 To active, check, or disable the hardware compression feature
1135 on an EXB-8900, use the exabyte MammothTool. You can get it here:
1136 \elink{http://www.exabyte.com/support/online/downloads/index.cfm}
1137 {http://www.exabyte.com/support/online/downloads/index.cfm}.
1138 There is a Solaris version of this tool. With option -C 0 or 1 you
1139 can disable or activate compression. Start this tool without any
1140 options for a small reference.
1143 \subsection{Using btape to Simulate Filling a Tape}
1144 \index[general]{Using btape to Simulate Filling a Tape}
1145 \index[general]{Tape!Using btape to Simulate Filling}
1147 Because there are often problems with certain tape drives or systems when end
1148 of tape conditions occur, {\bf btape} has a special command {\bf fill} that
1149 causes it to write random data to a tape until the tape fills. It then writes
1150 at least one more Bacula block to a second tape. Finally, it reads back both
1151 tapes to ensure that the data has been written in a way that Bacula can
1152 recover it. Note, there is also a single tape option as noted below, which you
1153 should use rather than the two tape test. See below for more details.
1155 This can be an extremely time consuming process (here it is about 6 hours) to
1156 fill a full tape. Note, that btape writes random data to the tape when it is
1157 filling it. This has two consequences: 1. it takes a bit longer to generate
1158 the data, especially on slow CPUs. 2. the total amount of data is
1159 approximately the real physical capacity of your tape, regardless of whether
1160 or not the tape drive compression is on or off. This is because random data
1161 does not compress very much.
1163 To begin this test, you enter the {\bf fill} command and follow the
1164 instructions. There are two options: the simple single tape option and the
1165 multiple tape option. Please use only the simple single tape option because
1166 the multiple tape option still doesn't work totally correctly. If the single
1167 tape option does not succeed, you should correct the problem before using
1169 \label{RecoveringFiles}
1171 \section{Recovering Files Written With Fixed Block Sizes}
1172 \index[general]{Recovering Files Written With Fixed Block Sizes}
1174 If you have been previously running your tape drive in fixed block mode
1175 (default 512) and Bacula with variable blocks (default), then in version
1176 1.32f-x and 1.34 and above, Bacula will fail to recover files because it does
1177 block spacing, and because the block sizes don't agree between your tape drive
1178 and Bacula it will not work.
1180 The long term solution is to run your drive in variable block mode as
1181 described above. However, if you have written tapes using fixed block sizes,
1182 this can be a bit of a pain. The solution to the problem is: while you are
1183 doing a restore command using a tape written in fixed block size, ensure that
1184 your drive is set to the fixed block size used while the tape was written.
1185 Then when doing the {\bf restore} command in the Console program, do not
1186 answer the prompt {\bf yes/mod/no}. Instead, edit the bootstrap file (the
1187 location is listed in the prompt) using any ASCII editor. Remove all {\bf
1188 VolBlock} lines in the file. When the file is re-written, answer the question,
1189 and Bacula will run without using block positioning, and it should recover
1193 \section{Tape Blocking Modes}
1194 \index[general]{Modes!Tape Blocking}
1195 \index[general]{Tape Blocking Modes}
1197 SCSI tapes may either be written in {\bf variable} or {\bf fixed} block sizes.
1198 Newer drives support both modes, but some drives such as the QIC devices
1199 always use fixed block sizes. Bacula attempts to fill and write complete
1200 blocks (default 65K), so that in normal mode (variable block size), Bacula
1201 will always write blocks of the same size except the last block of a Job. If
1202 Bacula is configured to write fixed block sizes, it will pad the last block of
1203 the Job to the correct size. Bacula expects variable tape block size drives to
1204 behave as follows: Each write to the drive results in a single record being
1205 written to the tape. Each read returns a single record. If you request less
1206 bytes than are in the record, only those number of bytes will be returned, but
1207 the entire logical record will have been read (the next read will retrieve the
1208 next record). Thus data from a single write is always returned in a single
1209 read, and sequentially written records are returned by sequential reads.
1211 Bacula expects fixed block size tape drives to behave as follows: If a write
1212 length is greater than the physical block size of the drive, the write will be
1213 written as two blocks each of the fixed physical size. This single write may
1214 become multiple physical records on the tape. (This is not a good situation).
1215 According to the documentation, one may never write an amount of data that is
1216 not the exact multiple of the blocksize (it is not specified if an error
1217 occurs or if the the last record is padded). When reading, it is my
1218 understanding that each read request reads one physical record from the tape.
1219 Due to the complications of fixed block size tape drives, you should avoid
1220 them if possible with Bacula, or you must be ABSOLUTELY certain that you use
1221 fixed block sizes within Bacula that correspond to the physical block size of
1222 the tape drive. This will ensure that Bacula has a one to one correspondence
1223 between what it writes and the physical record on the tape.
1225 Please note that Bacula will not function correctly if it writes a block and
1226 that block is split into two or more physical records on the tape. Bacula
1227 assumes that each write causes a single record to be written, and that it can
1228 sequentially recover each of the blocks it has written by using the same
1229 number of sequential reads as it had written.
1231 \section{Details of Tape Modes}
1232 \index[general]{Modes!Details}
1233 \index[general]{Details of Tape Modes}
1234 Rudolf Cejka has provided the following information concerning
1235 certain tape modes and MTEOM.
1239 It is always possible to position filemarks or blocks, whereas
1240 positioning to the end-of-data is only optional feature, however it is
1241 implemented very often. SCSI specification also talks about optional
1242 sequential filemarks, setmarks and sequential setmarks, but these are not
1243 implemented so often. Modern tape drives keep track of file positions in
1244 built-in chip (AIT, LTO) or at the beginning of the tape (SDLT), so there
1245 is not any speed difference, if end-of-data or filemarks is used (I have
1246 heard, that LTO-1 from all 3 manufacturers do not use its chip for file
1247 locations, but a tape as in SDLT case, and I'm not sure about LTO-2 and
1248 LTO-3 case). However there is a big difference, that end-of-data ignores
1249 file position, whereas filemarks returns the real number of skipped
1250 files, so OS can track current file number just in filemarks case.
1253 Solaris does use just SCSI SPACE Filemarks, it does not support SCSI
1254 SPACE End-of-data. When MTEOM is called, Solaris does use SCSI SPACE
1255 Filemarks with count = 1048576 for fast mode, and combination of SCSI
1256 SPACE Filemarks with count = 1 with SCSI SPACE Blocks with count = 1 for
1257 slow mode, so EOD mark on the tape on some older tape drives is not
1258 skipped. File number is always tracked for MTEOM.
1260 Linux does support both SCSI SPACE Filemarks and End-of-data: When MTEOM
1261 is called in MT\_ST\_FAST\_MTEOM mode, SCSI SPACE End-of-data is used.
1262 In the other case, SCSI SPACE Filemarks with count =
1264 There is no real slow mode like in Solaris - I just expect, that for
1265 older tape drives Filemarks may be slower than End-of-data, but not so
1266 much as in Solaris slow mode. File number is tracked for MTEOM just
1267 without MT\_ST\_FAST\_MTEOM - when MT\_ST\_FAST\_MTEOM is used, it is not.
1269 FreeBSD does support both SCSI SPACE Filemarks and End-of-data, but when
1270 MTEOD (MTEOM) is called, SCSI SPACE End-of-data is always used. FreeBSD
1271 never use SCSI SPACE Filemarks for MTEOD. File number is never tracked
1275 When {\bf Hardware End of Medium = Yes} is used, MTEOM is called, but it
1276 does not mean, that hardware End-of-data must be used. When Hardware End
1277 of Medium = No, if Fast Forward Space File = Yes, MTFSF with count =
1278 32767 is used, else Block Read with count = 1 with Forward Space File
1279 with count = 1 is used, which is really very slow.
1281 \item [Hardware End of Medium = Yes|No]
1282 The name of this option is misleading and is the source of confusion,
1283 because it is not the hardware EOM, what is really switched here.
1285 If I use Yes, OS must not use SCSI SPACE End-of-data, because Bacula
1286 expects, that there is tracked file number, which is not supported by
1287 SCSI specification. Instead, the OS have to use SCSI SPACE Filemarks.
1289 If I use No, an action depends on Fast Forward Space File.
1291 When I set {\bf Hardware End of Medium = no}
1292 and {\bf Fast Forward Space File = no}
1293 file positioning was very slow
1294 on my LTO-3 (about ten to 100 minutes), but
1296 with {\bf Hardware End of Medium = no} and
1297 {\bf Fast Forward Space File = yes}, the time is ten to
1298 100 times faster (about one to two minutes).
1302 \section{Tape Performance Problems}
1303 \index[general]{Tape Performance}
1304 If you have LTO-3 or LTO-4 drives, you should be able to
1305 fairly good transfer rates; from 60 to 150 MB/second, providing
1306 you have fast disks; GigaBit Ethernet connections (probably 2); you are
1307 running multiple simultaneous jobs; you have Bacula data spooling
1308 enabled; your tape block size is set to 131072 or 262144; and
1309 you have set {\bf Maximum File Size = 5G}.
1311 If you are not getting good performance, consider some of the following
1312 suggestions from the Allen Balck on the Bacula Users email list:
1315 \item You are using an old HBA (i.e. SCSI-1, which only does 5 MB/s)
1317 \item There are other, slower, devices on the SCSI bus. The HBA will
1318 negotiate the speed of every device down to the speed of the
1321 \item There is a termination problem on the bus (either too much or
1322 too little termination). The HBA will drop the bus speed in an
1323 attempt to increase the reliability of the bus.
1325 \item Loose or damaged cabling - this will probably make the HBA "think"
1326 you have a termination problem and it will react as in 3 above.
1329 See if /var/adm/messages (or /var/log/messages) tells you what the sync
1330 rate of the SCSI devices/bus are. Also, the next time you reboot, the
1331 BIOS may be able to tell you what the rate of each device is.
1334 \section{Autochanger Errors}
1335 \index[general]{Errors!Autochanger}
1336 \index[general]{Autochanger Errors}
1338 If you are getting errors such as:
1342 3992 Bad autochanger "load slot 1, drive 1": ERR=Child exited with code 1.
1346 and you are running your Storage daemon as non-root, then most likely
1347 you are having permissions problems with the control channel. Running
1348 as root, set permissions on /dev/sgX so that the userid and group of
1349 your Storage daemon can access the device. You need to ensure that you
1350 all access to the proper control device, and if you don't have any
1351 SCSI disk drives (including SATA drives), you might want to change
1352 the permissions on /dev/sg*.
1354 \section{Syslog Errors}
1355 \index[general]{Errors!Syslog}
1356 \index[general]{Syslog Errors}
1358 If you are getting errors such as:
1362 : kernel: st0: MTSETDRVBUFFER only allowed for root
1366 you are most likely running your Storage daemon as non-root, and
1367 Bacula is attempting to set the correct OS buffering to correspond
1368 to your Device resource. Most OSes allow only root to issue this
1369 ioctl command. In general, the message can be ignored providing
1370 you are sure that your OS parameters are properly configured as
1371 described earlier in this manual. If you are running your Storage daemon
1372 as root, you should not be getting these system log messages, and if
1373 you are, something is probably wrong.