Important: This documentation is about an older version. It's relevant only to the release noted, many of the features and functions have been updated or replaced. Please view the current version.
Upgrade your Tempo installation
You can upgrade an existing Tempo installation to the next version. However, any new release has the potential to have breaking changes that should be tested in a non-production environment prior to rolling these changes to production.
The upgrade process changes for each version, depending upon the changes made for the subsequent release.
This upgrade guide applies to on-premise installations and not for Grafana Cloud.
TIP: You can check your configuration options using the
status
API endpoint in your Tempo installation.
Upgrade to Tempo 2.3
Tempo 2.3 has several considerations for any upgrade:
- vParquet3 is available as a stable, production-read block format
- Configuration option to use Azure SDK v2
- New
defaults
block in Overrides module configuration - Several configuration parameters have been renamed or removed.
For a complete list of changes, enhancements, and bug fixes, refer to the Tempo 2.3 changelog.
Production-ready vParquet3 block format
vParquet3 provides improved query performance and dedicated attribute columns.
This block format is required for using dedicated attribute columns.
While vParquet2 remains the default backend for Tempo 2.3, vParquet3 is available as a stable option. Both work with Tempo 2.3.
Upgrading to Tempo 2.3 doesn’t modify the Parquet block format.
Note
Tempo 2.2 can’t read data stored in vParquet3.
Recommended update process:
- Upgrade your Tempo installation to version 2.3, remaining on vParquet2.
- Verify the upgrade is stable and performs as expected. If you notice any issues, you can downgrade to version 2.2, and data remains readable.
- Change the block format to vParquet3.
If you notice any issues on step 3 using the new block format, you can downgrade to vParquet2. All your data remains readable in Tempo 2.3. However, if you have vParquet3 blocks and have to downgrade to Tempo 2.2, you will have data loss.
Use Azure SDK v2
If you are using Azure storage, we recommend using the v2 SDK, azure-sdk-for-go.
You can use the use_v2_sdk
configure option for switching.
For more information, refer to the Storage block configuration example documentation.
New defaults
block in Overrides module configuration
The Overrides module has a new defaults
block for configuring global or per-tenant settings.
The Overrides format now includes changes to indented syntax.
For more information, read the Overrides configuration documentation.
You can also use the Tempo CLI to migrate configurations. Refer to the tempo-cli documentation.
The old configuration block looked like this:
overrides:
ingestion_rate_strategy: local
ingestion_rate_limit_bytes: 12345
ingestion_burst_size_bytes: 67890
max_search_duration: 17s
forwarders: ['foo']
metrics_generator_processors: [service-graphs, span-metrics]
The new configuration block looks like this:
overrides:
defaults:
ingestion:
rate_strategy: local
rate_limit_bytes: 12345
burst_size_bytes: 67890
read:
max_search_duration: 17s
forwarders: ['foo']
metrics_generator:
processors: [service-graphs, span-metrics]
Removed or renamed configuration parameters
Parameter | Comments |
distributor.log_received_traces | Use the distributor.log_received_spans configuration block instead. [PR #3008] |
tempo_query_frontend_queries_total{op="searchtags|metrics"} | Removed deprecated frontend metrics configuration option |
Upgrade to Tempo 2.2
Tempo 2.2 has several considerations for any upgrade:
- vParquet2 is now the default block format
- Several configuration parameters have been renamed or removed.
For a complete list of changes, enhancements, and bug fixes, refer to the Tempo 2.2 changelog.
Default block format changed to vParquet2
While not a breaking change, upgrading to Tempo 2.2 will by default change Tempo’s block format to vParquet2.
To stay on a previous block format, read the Parquet configuration documentation.
We strongly encourage upgrading to vParquet2 as soon as possible as this is required for using structural operators in your TraceQL queries and provides query performance improvements, in particular on queries using the duration
intrinsic.
Updated JSonnet supports statefulset
for the metrics-generator
Tempo 2.2 updates the microservices
JSonnet to support a statefulset
for the metrics_generator
component.
Note
This update is important if you use the experimentallocal-blocks
processor.
To support a new processor
, the metrics-generator has been converted from a deployment
into a statefulset
with a PVC.
This requires manual intervention to migrate successfully and avoid downtime.
Note that currently both a deployment
and a statefulset
will be managed by the JSonnet for a period of time, after which we will delete the deployment from this repo and you will need to delete user-side references to the tempo_metrics_generator_deployment
, as well as delete the deployment itself.
Refer to the PR for seamless migration instructions. [PRs 2533, 2467]
Removed or renamed configuration parameters
The following fields were removed or renamed.
Parameter | Comments |
---|---|
query_frontend: | Remove support for tolerant_failed_blocks [PR 2416] |
storage: | Renamed insecure_skip_verify to tls_insecure_skip_verify [PR 2407] |
Upgrade to Tempo 2.1
Tempo 2.1 has two major considerations for any upgrade:
- Support for search on v2 block is removed
- Breaking changes to metric names
For more information on other enhancements, read the Tempo 2.1 release notes.
Remove support for Search on v2 blocks
Users can no longer search blocks in v2 format. Only vParquet and vParquet2 formats support search. The following search configuration options were removed from the overrides section:
overrides:
max_search_bytes_per_trace:
search_tags_allow_list:
search_tags_deny_list:
The following metrics configuration was also removed:
tempo_ingester_trace_search_bytes_discarded_total
Upgrade path to maintain search from Tempo 1.x to 2.1
Removing support for search on v2 blocks means that if you upgrade directly from 1.9 to 2.1, you will not be able to search your v2 blocks. To avoid this, upgrade to 2.0 first, since 2.0 supports searching both v2 and vParquet blocks. You can let your old v2 blocks gradually age out while Tempo creates new vParquet blocks from incoming traces. Once all of your v2 blocks have been deleted and you only have vParquet format-blocks, you can upgrade to Tempo 2.1. All of your blocks will be searchable.
Parquet files are no longer cached when carrying out searches.
Breaking changes to metric names exposed by Tempo
All Prometheus metrics exposed by Tempo on its /metrics
endpoint that were previously prefixed with cortex_
have now been renamed to be prefixed with tempo_
instead. (PR 2204)
Tempo now includes SLO metrics to count where queries are returned within a configurable time range. (PR 2008)
The query_frontend_result_metrics_inspected_bytes
metric was removed in favor of query_frontend_bytes_processed_per_second
.
Upgrade from Tempo 1.5 to 2.0
Tempo 2.0 marks a major milestone in Tempo’s development. When planning your upgrade, consider these factors:
- Breaking changes:
- Renamed, removed, and moved configurations are described in section below.
- The
TempoRequestErrors
alert was removed from mixin. Any Jsonnet users relying on this alert should copy this into their own environment.
- Advisory:
- Changed defaults – Are these updates relevant for your installation?
- TraceQL editor needs to be enabled in Grafana to use the query editor.
- Resource requirements have changed for Tempo 2.0 with the default configuration.
Once you upgrade to Tempo 2.0, there is no path to downgrade.
Note
There is a potential issue loading Tempo 1.5’s experimental Parquet storage blocks. You may see errors or even panics in the compactors. We have only been able to reproduce this with interim commits between 1.5 and 2.0, but if you experience any issues please report them so we can isolate and fix this issue.
Check Tempo installation resource allocation
Parquet provides faster search and is required to enable TraceQL. However, the Tempo installation will require additional CPU and memory resources to use Parquet efficiently. Parquet is more costly due to the extra work of building the columnar blocks, and operators should expect at least 1.5x increase in required resources to run a Tempo 2.0 cluster. Most users will find these extra resources are negligible compared to the benefits that come from the additional features of TraceQL and from storing traces in an open format.
You can can continue using the previous v2
block format using the instructions provided in the Parquet configuration documentation. Tempo will continue to support trace by id lookup on the v2
format for the foreseeable future.
Enable TraceQL in Grafana
TraceQL is enabled by default in Tempo 2.0. The TraceQL query editor requires Grafana 9.3.2 and later.
The TraceQL query editor is in beta in Grafana 9.3.2 and needs to be enabled with the traceqlEditor
feature flag.
Check configuration options for removed and renamed options
The following tables describe the parameters that have been removed or renamed.
Removed and replaced
Parameter | Comments |
---|---|
query_frontend: | Replaced by trace_by_id.query_shards . |
querier: | Replaced by two different settings: search.query_timeout and trace_by_id.query_timeout . |
ingester: | Removed and automatically determined based on block format. |
search_enabled | Removed. Now defaults to true. |
metrics_generator_enabled | Removed. Now defaults to true. |
Renamed
The following compactor
configuration parameters were renamed.
Parameter | Comments |
---|---|
compaction: | Renamed to v2_in_buffer_bytes |
compaction: | Renamed to v2_out_buffer_bytes |
compaction: | Renamed to v2_prefetch_traces_count |
The following storage
configuration parameters were renamed.
Parameter | Comments |
---|---|
wal: | Renamed to v2_encoding |
block: | Renamed to v2_index_downsample_bytes |
block: | Renamed to v2_index_page_size_bytes |
block: | Renamed to v2_encoding |
block: | Renamed to parquet_row_group_size_bytes |
The Azure Storage configuration section now uses snake case with underscores (_
) instead of dashes (-
). Example of using snake case on Azure Storage config:
# config.yaml
storage:
trace:
azure:
storage_account_name:
storage_account_key:
container_name: