Configure heap dump path for archive packages (#29130)

This is a follow up to a previous change which set the heap dump path
for the package distributions. The observation here is that we always
set the working directory of Elasticsearch to to the root of
installation (i.e., Elasticsearch home). Therefore, we can specify the
heap dump path relative to this directory and default it to the data
directory, similar to the package distributions.
This commit is contained in:
Jason Tedor 2018-03-18 15:32:37 -04:00 committed by GitHub
parent 1d8c507684
commit 29fedb2669
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
3 changed files with 15 additions and 19 deletions

View File

@ -262,7 +262,7 @@ subprojects {
'heap.dump.path': [
'deb': "-XX:HeapDumpPath=/var/lib/elasticsearch",
'rpm': "-XX:HeapDumpPath=/var/lib/elasticsearch",
'def': "#-XX:HeapDumpPath=/heap/dump/path"
'def': "-XX:HeapDumpPath=data"
],
'error.file': [

View File

@ -77,8 +77,8 @@
# heap dumps are created in the working directory of the JVM
-XX:+HeapDumpOnOutOfMemoryError
# specify an alternative path for heap dumps
# ensure the directory exists and has sufficient space
# specify an alternative path for heap dumps; ensure the directory exists and
# has sufficient space
${heap.dump.path}
# specify an alternative path for JVM fatal error logs

View File

@ -1,19 +1,15 @@
[[heap-dump-path]]
=== JVM heap dump path
The <<rpm,RPM>> and <<deb,Debian>> package distributions default to configuring
the JVM to dump the heap on out of memory exceptions to
`/var/lib/elasticsearch`. If this path is not suitable for storing heap dumps,
you should modify the entry `-XX:HeapDumpPath=/var/lib/elasticsearch` in
<<jvm-options,`jvm.options`>> to an alternate path. If you specify a filename
instead of a directory, the JVM will repeatedly use the same file; this is one
mechanism for preventing heap dumps from accumulating in the heap dump path.
Alternatively, you can configure a scheduled task via your OS to remove heap
dumps that are older than a configured age.
Note that the archive distributions do not configure the heap dump path by
default. Instead, the JVM will default to dumping to the working directory for
the Elasticsearch process. If you wish to configure a heap dump path, you should
modify the entry `#-XX:HeapDumpPath=/heap/dump/path` in
<<jvm-options,`jvm.options`>> to remove the comment marker `#` and to specify an
actual path.
By default, Elasticsearch configures the JVM to dump the heap on out of
memory exceptions to the default data directory (this is
`/var/lib/elasticsearch` for the <<rpm,RPM>> and <<deb,Debian>> package
distributions, and the `data` directory under the root of the
Elasticsearch installation for the <<zip-targz,tar and zip>> archive
distributions). If this path is not suitable for receiving heap dumps,
you should modify the entry `-XX:HeapDumpPath=...` in
<<jvm-options,`jvm.options`>>. If you specify a fixed filename instead
of a directory, the JVM will repeatedly use the same file; this is one
mechanism for preventing heap dumps from accumulating in the heap dump
path. Alternatively, you can configure a scheduled task via your OS to
remove heap dumps that are older than a configured age.