f2e4c40cee
* refactor toc * refactor toc * Change to pydata-sphinx-theme and update packages requirement list for ReadtheDocs * Remove customized css for old theme * Add index page to each top bar section and limit dropdown maximum to be 4 * Use js to change 'More' to 'Libraries' * Add custom.css to conf.py for further css changes * Add BigDL logo and search bar * refactor toc * refactor toc and add overview * refactor toc and add overview * refactor toc and add overview * refactor get started * add paper and video section * add videos * add grid columns in landing page * add document roadmap to index * reapply search bar and github icon commit * reorg orca and chronos sections * Test: weaken ads by js * update: change left attrbute * update: add comments * update: change opacity to 0.7 * Remove useless theme template override for old theme * Add sidebar releases component in the home page * Remove sidebar search and restore top nav search button * Add BigDL handouts * Add back to homepage button to pages except from the home page * Update releases contents & styles in left sidebar * Add version badge to the top bar * Test: weaken ads by js * update: add comments * remove landing page contents * rfix chronos install * refactor install * refactor chronos section titles * refactor nano index * change chronos landing * revise chronos landing page * add document navigator to nano landing page * revise install landing page * Improve css of versions in sidebar * Make handouts image pointing to a page in new tab * add win guide to install * add dliib installation * revise title bar * rename index files * add index page for user guide * add dllib and orca API * update user guide landing page * refactor side bar * Remove extra style configuration of card components & make different card usage consistent * Remove extra styles for Nano how-to guides * Remove extra styles for Chronos how-to guides * Remove dark mode for now * Update index page description * Add decision tree for choosing BigDL libraries in index page * add dllib models api, revise core layers formats * Change primary & info color in light mode * Restyle card components * Restructure Chronos landing page * Update card style * Update BigDL library selection decision tree * Fix failed Chronos tutorials filter * refactor PPML documents * refactor and add friesian documents * add friesian arch diagram * update landing pages and fill key features guide index page * Restyle link card component * Style video frames in PPML sections * Adjust Nano landing page * put api docs to the last in index for convinience * Make badge horizontal padding smaller & small changes * Change the second letter of all header titles to be small capitalizd * Small changes on Chronos index page * Revise decision tree to make it smaller * Update: try to change the position of ads. * Bugfix: deleted nonexist file config * Update: update ad JS/CSS/config * Update: change ad. * Update: delete my template and change files. * Update: change chronos installation table color. * Update: change table font color to --pst-color-primary-text * Remove old contents in landing page sidebar * Restyle badge for usage in card footer again * Add quicklinks template on landing page sidebar * add quick links * Add scala logo * move tf, pytorch out of the link * change orca key features cards * fix typo * fix a mistake in wording * Restyle badge for card footer * Update decision tree * Remove useless html templates * add more api docs and update tutorials in dllib * update chronos install using new style * merge changes in nano doc from master * fix quickstart links in sidebar quicklinks * Make tables responsive * Fix overflow in api doc * Fix list indents problems in [User guide] section * Further fixes to nested bullets contents in [User Guide] section * Fix strange title in Nano 5-min doc * Fix list indent problems in [DLlib] section * Fix misnumbered list problems and other small fixes for [Chronos] section * Fix list indent problems and other small fixes for [Friesian] section * Fix list indent problem and other small fixes for [PPML] section * Fix list indent problem for developer guide * Fix list indent problem for [Cluster Serving] section * fix dllib links * Fix wrong relative link in section landing page Co-authored-by: Yuwen Hu <yuwen.hu@intel.com> Co-authored-by: Juntao Luo <1072087358@qq.com>
53 lines
2.1 KiB
Markdown
53 lines
2.1 KiB
Markdown
# Cluster Serving FAQ
|
|
|
|
## General Debug Guide
|
|
You could use following guide to debug if serving is not working properly.
|
|
|
|
### Check if Cluster Serving environment is ready
|
|
Run following commands in terminal
|
|
```
|
|
echo $FLINK_HOME
|
|
echo $REDIS_HOME
|
|
```
|
|
the output directory
|
|
```
|
|
/path/to/flink-version
|
|
/path/to/redis-version
|
|
```
|
|
|
|
should be displayed, otherwise, go to [Programming Guide](ProgrammingGuide.md) **Installation** section.
|
|
|
|
### Check if Flink Cluster is working
|
|
Run following commands in terminal
|
|
```
|
|
netstat -tnlp
|
|
```
|
|
output like following should be displayed, `6123,8081` is Flink default port usage.
|
|
```
|
|
tcp6 0 0 :::6123 :::* LISTEN xxxxx/java
|
|
tcp6 0 0 :::8081 :::* LISTEN xxxxx/java
|
|
```
|
|
if not, run `$FLINK_HOME/bin/start-cluster.sh` to start Flink cluster.
|
|
|
|
After that, check Flink log in `$FLINK_HOME/log/`, check the log file of `flink-xxx-standalone-xxx.log` and `flink-xxx-taskexecutor-xxx.log` to make sure there is no error.
|
|
|
|
If the port could not bind in this step, kill the program which use the port, and `$FLINK_HOME/bin/stop-cluster.sh && $FLINK_HOME/bin/start-cluster.sh` to restart Flink cluster.
|
|
### Check if Cluster Serving is running
|
|
```
|
|
$FLINK_HOME/bin/flink list
|
|
```
|
|
output of Cluster Serving job information should be displayed, if not, go to [Programming Guide](ProgrammingGuide.md) **Launching Service** section to make sure you call `cluster-serving-start` correctly.
|
|
|
|
|
|
|
|
### Troubleshooting
|
|
|
|
1. `Duplicate registration of device factory for type XLA_CPU with the same priority 50`
|
|
|
|
This error is caused by Flink ClassLoader. Please put cluster serving related jars into `${FLINK_HOME}/lib`.
|
|
|
|
2. `servable Manager config dir not exist`
|
|
|
|
Check if `servables.yaml` exists in current directory. If not, download from [github](https://github.com/intel-analytics/bigdl/blob/master/ppml/trusted-realtime-ml/scala/docker-graphene/servables.yaml).
|
|
### Still, I get no result
|
|
If you still get empty result, raise issue [here](https://github.com/intel-analytics/bigdl/issues) and post the output/log of your serving job.
|