1 Observer Release Notes

This document describes the changes made to the Observer application.

1.1 Observer 1.3

Improvements and New Features

Where necessary a comment stating encoding has been added to Erlang files. The comment is meant to be removed in Erlang/OTP R17B when UTF-8 becomes the default encoding.

Own Id: OTP-10630

1.2 Observer 1.2

Fixed Bugs and Malfunctions

The module name in the link to the detail page for each loaded module was earlier not URL encoded. If the module name contained e.g. a # this could cause the crashdump viewer to crash when opening the link. This has been corrected.

Own Id: OTP-10090 Aux Id: seq12068
Escape control characters in Table Viewer

Similar behaviour to old tv. Objects in tables supposed to be printed in a single line and it looks ugly when a [...,10,...] integer list creates a new-line. Fix Table Viewer search crash on new|changed|deleted rows.

Fix Table Viewer crash after a 'Found' -> 'Not found' search sequence

Start position was lost after a 'Found' -> 'Not found' search sequence leading an undefined position in the next search. Thanks to Peti Gömori

Own Id: OTP-10218
observer: fix app file (Noticed-by: Motiejus Jakstys)

Add missing observer modules to observer.app.src. Thanks to Tuncer Ayaz.

Own Id: OTP-10221
Make Table Viewer search a bit faster

Edit table row in a multiline text dialog. Thanks to Peti Gomori.

Own Id: OTP-10225

Improvements and New Features

Allow tracing on bifs.

Ask epmd for local nodes, and remember users last input in connect.

Fix crashes when a table or process information could not be retrieved.

Own Id: OTP-10075

1.3 Observer 1.0

Fixed Bugs and Malfunctions

The following bugs in ttb have been corrected:
- ttb:tracer/2 would earlier crash when trying to set up tracing for a diskless node to wrap files, i.e. when option {file,{local,{wrap,Filename,Size,Count}}} was used.
- ttb:stop([fetch]) would sometimes silently fail if multiple nodes with different current working directories were traced.
- ttb:stop([fetch]) would crash if the tracer was started with option {file,{local,Filename}}
- A deadlock would sometimes occur due to an information printout from the ttb_control process when ttb was stopped.
Own Id: OTP-9431
The file trace port to which the IP trace client relays all traces from diskless nodes was not flushed and closed properly on ttb:stop. This has been corrected.

Own Id: OTP-9665

Improvements and New Features

A new GUI for Observer. Integrating pman, etop, appmon and tv into observer with tracing facilities.

Own Id: OTP-4779
The following new features are added to ttb:
- A one-command trace setup is added, ttb:start_trace/4.
- The following new options are added to ttb:tracer/2:
  - shell - Show trace messages on the console in real time
  - timer - Time constrained tracing
  - overload - Overload protection
  - flush - Flush file trace port buffers with given frequency
  - resume - Automatically resume tracing after node restart
- A new shortcut is added for common tracer settings similar to using the dbg module directly, ttb:tracer(shell | dbg).
- Some shortcuts are added for commonly used match specifications in ttb:tp and ttb:tpl.
- The Options argument to functions ttb:tracer, ttb:write_config, ttb:stop and ttb:format may now be one single option instead of a list.
- The history buffer of the last trace is now always automatically dumped to the file ttb_last_config when ttb:stop is called.
- The following new options are added to ttb:stop/1:
  - fetch_dir - Specify where to store fetched logs
  - {format,FormatOpts} - Specify options to use when formatting the fetched logs
  - return_fetch_dir - Indicate that the return value from ttb:stop/1 should include the name of the directory where the fetched logs are stored
- The option disable_sort is added to ttb:format/2. When this option is used, trace messages from different logs are not merged according to timestamps, but just appended one log after the other.
Own Id: OTP-9403
The following non backwards compatible changes are done in ttb:
- When setting up trace with ttb, the 'timestamp' trace flag will now always be set.
- The 'fetch' option to ttb:stop/1 is removed since it is now default behavior that trace logs are fetched when stopping ttb. Fetching can be disabled with the 'nofetch' option to ttb:stop/1.
- The name of the upload directory is changed from ttb_upload-Timestamp to ttb_upload_FileName-Timestamp.
- To format the output using 'et', you now need to provide the option {handler,ttb:get_et_handler()} instead of {handler,et}.
- When formatting a trace log, the handler state was earlier reset after each trace file, this is now changed so the handler state is passed not only from one trace message to the next in the same file, but also from one file to the next.
*** POTENTIAL INCOMPATIBILITY ***

Own Id: OTP-9430

1.4 Observer 0.9.10

Fixed Bugs and Malfunctions

Do not install *.bat files on non-win32 machines (Thanks to Hans Ulrich Niedermann)

Own Id: OTP-9515

1.5 Observer 0.9.9

Improvements and New Features

The time needed for loading a crashump into the crashdump viewer would earlier grow exponentially with the size of the crashdump file. Reading a file of 20M would take a couple of minutes, and for a dump of 250M it would take between 1 and 2 hours. This has been solved.

Earlier, all processes, timers, funs or ets-tables would be loaded into the memory of the crashdump viewer node before sending it on to the web server. This has been changed and the pages are now sent to the web server in chunks.

A security function in newer web browsers prevents a full file path to be sent from an HTML file input field, i.e. the field needed to implement the "Browse" button when loading a file into the crashdump viewer. To overcome this, the file input field is no longer used. Instead a normal text input field is used, and the user needs to manually insert the complete file path. For convenience, a shell script and a batch file are added to the observer application. These can be used to start the crashdump_viewer and a browser and load a file - with the file name given from the command line. The shell script and batch file are called cdv and cdv.bat respectively, and can be found in the priv dir of the observer application.

Own Id: OTP-9051 Aux Id: seq11789

1.6 Observer 0.9.8.4

Improvements and New Features

The multitrace.erl installation example file is now installed in the examples directory. (Thanks to Peter Lemenkov.)

Own Id: OTP-8857

1.7 Observer 0.9.8.3

Improvements and New Features

The test suite has been updated for R14A.

Own Id: OTP-8708

1.8 Observer 0.9.8.2

Improvements and New Features

Misc updates

Own Id: OTP-8456

1.9 Observer 0.9.8.1

Improvements and New Features

Major improvements of the Erlang distribution for Erlang runtime systems with SMP support. Previously distribution port locks were heavily contended, and all encoding and decoding for a specific distribution channel had to be done in sequence. Lock contention due to the distribution is now negligible and both encoding and decoding of Erlang messages can be done in parallel.

The old atom cache protocol used by the Erlang distribution has been dropped since it effectively prevented all parallel encoding and decoding of messages passed over the same distribution channel.

A new atom cache protocol has been introduced which isolates atom cache accesses and makes parallel encoding and decoding of messages passed over the same distribution channel possible. The new atom cache protocol also use an atom cache size 8 times larger than before. The new atom cache protocol is documented in the ERTS users guide.

Erlang messages received via the distribution are now decoded by the receiving Erlang processes without holding any distribution channel specific locks. Erlang messages and signals sent over the distribution are as before encoded by the sending Erlang process, but now without holding any distribution channel specific locks during the encoding. That is, both encoding and decoding can be and are done in parallel regardless of distribution channel used.

The part that cannot be parallelized is the atom cache updates. Atom cache updates are therefore now scheduled on the distribution port. Since it is only one entity per distribution channel doing this work there is no lock contention due to the atom cache updates.

The new runtime system does not understand the old atom cache protocol. New and old runtime systems can however still communicate, but no atom cache will be used.

Own Id: OTP-7774

1.10 Observer 0.9.8

Improvements and New Features

etop would crash if the emulator's custom allocators had been turned off (e.g. using the +Meamin option).

Own Id: OTP-7519
The copyright notices have been updated.

Own Id: OTP-7851

1.11 Observer 0.9.7.4

Improvements and New Features

Minor Makefile changes.

Own Id: OTP-6689
Obsolete guard tests (such as list()) have been replaced with the modern guard tests (such as is_list()).

Own Id: OTP-6725

1.12 Observer 0.9.7.3

Improvements and New Features

This application has been updated to eliminate warnings by Dialyzer.

Own Id: OTP-6551

1.13 Observer 0.9.7.2

Fixed Bugs and Malfunctions

Several minor bugs and race conditions eliminated in the runtime_tools and observer applications.

Own Id: OTP-6265

1.14 Observer 0.9.7.1

Fixed Bugs and Malfunctions

Crash dump with large integers could crash the crashdump_viewer.

Own Id: OTP-6301

1.15 Observer 0.9.7 (R11B)

Fixed Bugs and Malfunctions

Fixed a bug in etop which made the Load and Memory information in the header incorrect -- for example the value shown for binary (memory allocated for binaries) was actually the number of currently running processes. (Thanks to Rikard Johansson.)

Own Id: OTP-6075

1.16 Observer 0.9.6.2

Fixed Bugs and Malfunctions

The Observer application has been recompiled because of a compiler bug.

Own Id: OTP-5700

1.17 Observer 0.9.6.1

Fixed Bugs and Malfunctions

crashdump_viewer is faster when showing message, dictionary, and stack dump for large processes.

Own Id: OTP-5408