public inbox for pbs-devel@lists.proxmox.com
 help / color / mirror / Atom feed
From: Stefan Reiter <s.reiter@proxmox.com>
To: Thomas Lamprecht <t.lamprecht@proxmox.com>,
	Proxmox Backup Server development discussion
	<pbs-devel@lists.proxmox.com>
Subject: Re: [pbs-devel] [PATCH proxmox-backup 1/5] backup: don't validate chunk existance if base was recently verified
Date: Wed, 30 Sep 2020 15:42:50 +0200	[thread overview]
Message-ID: <b18f71e0-004b-0e10-6af9-27fe80cbf9ac@proxmox.com> (raw)
In-Reply-To: <102da918-162c-fcf4-f1e7-68b46b63f953@proxmox.com>

On 9/30/20 3:32 PM, Thomas Lamprecht wrote:
> On 30.09.20 15:25, Stefan Reiter wrote:
>> If the base was successfully verified within the last 7 days, we assume
>> that it is okay and all chunks exist, so we don't have to check.
>>
>> Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
>> ---
>>   src/api2/backup/environment.rs | 29 ++++++++++++++++++++++++++++-
>>   1 file changed, 28 insertions(+), 1 deletion(-)
>>
>> diff --git a/src/api2/backup/environment.rs b/src/api2/backup/environment.rs
>> index d515bf30..be06d1dc 100644
>> --- a/src/api2/backup/environment.rs
>> +++ b/src/api2/backup/environment.rs
>> @@ -457,6 +457,31 @@ impl BackupEnvironment {
>>           Ok(())
>>       }
>>   
>> +    fn last_backup_has_recent_verify(&self) -> Result<bool, Error> {
>> +        match &self.last_backup {
>> +            Some(last_backup) => {
>> +                let last_dir = &last_backup.backup_dir;
>> +                let (manifest, _) = self.datastore.load_manifest(last_dir)?;
>> +                let verify = manifest.unprotected["verify_state"].clone();
>> +                match serde_json::from_value::<Option<SnapshotVerifyState>>(verify) {
>> +                    Ok(verify) => match verify {
>> +                        Some(verify) => {
>> +                            let cutoff = unsafe { libc::time(std::ptr::null_mut()) };
>> +                            let cutoff = cutoff - 60*60*24*7; // one week back
> 
> Why unsafe and why not our `proxmox::tools::time::epoch_i64()` ?
> 

Didn't know that exists, I've seen that unsafe a few times so I thought 
that was the usual way. I'll send v2.

> 
> on another note: we should probably add some helper for getting the
> verify state
> 

I'll see if it's feasible. So far we only have two sites that both do 
different things on individual errors, so I'm not sure how much that 
would simplify things.

>> +                            Ok(verify.state == VerifyState::Ok && verify.upid.starttime > cutoff)
>> +                        },
>> +                        None => Ok(false)
>> +                    },
>> +                    Err(err) => {
>> +                        self.worker.warn(format!("error parsing base verification state : '{}'", err));
>> +                        Ok(false)
>> +                    }
>> +                }
>> +            },
>> +            None => Ok(false)
>> +        }
>> +    }
>> +
>>       /// Ensure all chunks referenced in this backup actually exist.
>>       /// Only call *after* all writers have been closed, to avoid race with GC.
>>       /// In case of error, mark the previous backup as 'verify failed'.
>> @@ -534,7 +559,9 @@ impl BackupEnvironment {
>>               }
>>           }
>>   
>> -        self.verify_chunk_existance(&state.known_chunks)?;
>> +        if !self.last_backup_has_recent_verify()? {
>> +            self.verify_chunk_existance(&state.known_chunks)?;
>> +        }
>>   
>>           // marks the backup as successful
>>           state.finished = true;
>>
> 
> 




  reply	other threads:[~2020-09-30 13:43 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-09-30 13:25 [pbs-devel] [PATCH 0/5] backup validation improvements Stefan Reiter
2020-09-30 13:25 ` [pbs-devel] [PATCH proxmox-backup 1/5] backup: don't validate chunk existance if base was recently verified Stefan Reiter
2020-09-30 13:32   ` Thomas Lamprecht
2020-09-30 13:42     ` Stefan Reiter [this message]
2020-09-30 13:55       ` Thomas Lamprecht
2020-09-30 13:25 ` [pbs-devel] [RFC proxmox-backup 2/5] ParallelHandler: add unbounded mode Stefan Reiter
2020-09-30 13:25 ` [pbs-devel] [RFC proxmox-backup 3/5] ParallelHandler: add check_abort function and handle errors during join Stefan Reiter
2020-09-30 13:25 ` [pbs-devel] [RFC proxmox-backup 4/5] ParallelHandler: exit early if this or other thread aborted Stefan Reiter
2020-09-30 13:25 ` [pbs-devel] [RFC proxmox-backup 5/5] backup: validate chunk existance in background Stefan Reiter

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=b18f71e0-004b-0e10-6af9-27fe80cbf9ac@proxmox.com \
    --to=s.reiter@proxmox.com \
    --cc=pbs-devel@lists.proxmox.com \
    --cc=t.lamprecht@proxmox.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
Service provided by Proxmox Server Solutions GmbH | Privacy | Legal